Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Next Article in Journal
Multi-Step Ahead Forecasting of the Energy Consumed by the Residential and Commercial Sectors in the United States Based on a Hybrid CNN-BiLSTM Model
Previous Article in Journal
School-Based Interventions for Migrant Students in the Framework of the Health Promoting Whole-School Approach: An Umbrella Review
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 15
Issue 3
10.3390/su15031896
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

How Does Digital Transformation Facilitate Enterprise Total Factor Productivity? The Multiple Mediators of Supplier Concentration and Customer Concentration

by
Hua Zhang
1 and
Qiwang Zhang
2,*
1
Sunwah International Business School, Faculty of Economics, Liaoning University, Shenyang 100136, China
2
Business School, Faculty of Economics, Liaoning University, Shenyang 100136, China
*
Author to whom correspondence should be addressed.
Sustainability 2023, 15(3), 1896; https://doi.org/10.3390/su15031896
Submission received: 1 November 2022 / Revised: 10 January 2023 / Accepted: 12 January 2023 / Published: 19 January 2023
Browse Figures
Versions Notes

Abstract

:
Nowadays, no organization or sector is immune to the influences of digital transformation. It has been generally recognized as a significant promoter of corporate efficiency and effectiveness. However, the literature on whether and how digital transformation can facilitate corporate total factor productivity is scarce. In this case, this paper aims to empirically investigate whether and how digital transformation can boost a firm’s total factor productivity. Using a sample of Chinese listed companies from 2007–2020 and structural equation model analyses in STATA 16.0, this paper identifies that (1) digital technology, one dimension of digital transformation, has no significant effects on enterprise total factor productivity, while digital application, the other dimension of digital transformation, can directly enhance enterprise total factor productivity; (2) both digital technology and digital application can indirectly boost enterprise total factor productivity through the decreasing of supplier concentration; and (3) digital technology adversely affects enterprise total factor productivity through the increasing of customer concentration, and digital application may positively affect enterprise total factor productivity through the decreasing of customer concentration. Our study is one of the first to explore the mediating effect mechanism of the relationship between digital transformation and enterprise total factor productivity from the perspective of supply chain management.

1. Introduction

The development of the digital economy has brought unprecedented disruptions to society, industry and organizations [1,2]. New business opportunities associated with Industry 4.0 are emerging, which means that enterprises must adapt to the new environment to achieve sustainable development [3,4]. In recent years, governments around the world have been constantly issuing new policies to encourage the digital transformation of legacy enterprises, which is universally accepted as a significant promotor of efficiency and effectiveness [5,6]. Under the circumstances, according to the latest data from the China Digital Development Indicator Report 2021 (available at: https://www.xdyanbao.com/doc/l3i2fuk0al, accessed on 8 December 2022), China’s digital economy development index showed an upward trend year by year from 2012 to 2020. Compared with the benchmark index of 1000 in 2012, the digital economy development index increased to 2810.75 in 2020, with a growth rate of 181.075% (Figure 1a). This illustrated that the overall development levels of digital industrialization, industrial digitization and digital technology in China were continuously growing up. Meanwhile, the total factor productivity of the listed digital enterprises in China gradually increased from 4.5126 in 2012 to 5.0931 in 2020 (Figure 1b). It seems that digital transformation of enterprises has driven the improvement of corporate total factor productivity.
Nevertheless, in the academic field, research on the relationship between digital transformation and corporate total factor productivity is still in its infancy. Among those limited studies, scholars generally concluded that digital transformation can directly facilitate total factor productivity. For instance, Pan et al. [7] empirically verified that digital economy index has a positively nonlinear relationship with provincial total factor productivity, which means that the digital economy can act as an innovation driver for the extensive and sustainable development of total factor productivity. Similarly, Zeng and Lei [8] empirically identified that corporate digital transformation significantly enhances total factor productivity by improving the management efficiency and technical level of enterprises. These studies focused on whether digital transformation can boost total factor productivity (direct effect), while studies on how digital transformation can promote total factor productivity (indirect effect mechanism) were relatively rare. Although such research was rare, scholars have carried out a series of empirical studies, verifying that technological progress can dramatically promote green total factor productivity directly and indirectly [9,10,11,12,13,14]. In addition, scholars have empirically verified the positive effects of digital transformation on corporate innovation and productivity [4,15,16,17]. These studies, to some extent, provide some theoretical basis and statistical methodology for our further research on the relationship between digital transformation and enterprise total factor productivity.
Accordingly, based on resource-based theory [18,19], dynamic capability theory [20,21], industrial competitive theory [22] and transaction cost economics theory [23,24], this paper aims to investigate the possible mediating roles of customer concentration and supplier concentration in the relationship between digital transformation and total factor productivity. Specifically, this paper attempts to examine the following research questions:
  • Does digital transformation directly facilitate enterprise total factor productivity?
  • How does digital transformation indirectly boost enterprise total factor productivity through the decreasing of supplier concentration?
  • How does digital transformation indirectly enhance enterprise total factor productivity through the increasing of customer concentration?
In order to deal with these research questions, we adopt a quantitative approach using equation model analyses in STATA 16.0 and analyzing the data of 21,799 firm-year observations from 2007–2020 in China. On a basis of the extant literature [25,26,27], we defined digital transformation as the integration of digital technologies and applications into business processes. Accordingly, digital transformation relies on two dimensions, digital technology, which refers to all fundamental technologies with digital elements such as artificial intelligence, blockchains, cloud computing, big data technology, etc. [28]; and digital application, which includes the upgraded utilization of digital technology to generate effective outputs and applications, such as smart wear, mobile payment, unmanned retail, etc. [29]. Our empirical results indicated that (1) digital technology has no direct effects on enterprise total factor productivity, while digital application is directly positively related to total factor productivity; (2) both digital technology and digital application can indirectly facilitate enterprise total factor productivity through the decreasing of supplier concentration; and (3) differently from our proposed hypothesis, digital technology probably increases customer concentration but adversely affects enterprise total factor productivity; digital application may positively affect enterprise total factor productivity, but it does so through the decreasing of customer concentration.
Based on our findings, this paper contributes to the extant literature in the following aspects: first, we divide digital transformation into digital technology and digital application to conduct in-depth analysis on the effects of digital transformation on enterprise total factor productivity, which will provide guidance for subsequent research on digital transformation. Second, our paper is one of the first to explore the indirect mediating effect of supplier concentration and customer concentration on the relationship between digital transformation and enterprise total factor productivity. We provide a new perspective, based on the perspective of supply chain management, to explore the mediating effect mechanism. Despite these contributions, this paper has its limitations. The sample data were collected in China, and we need to consider whether the findings are still tenable and applicable in other national contexts. The intensity of digital transformation is measured by word frequency statistics for more than 80 keywords. Due to the rapid development of digital economy, we cannot cover all the keywords related to digital transformation in real time. We can collect more data in other countries and expand our keyword selection scope related to digital transformation in future research.
This paper is structured as follows: in the next section, we introduce the theoretical basis and proposed hypotheses. Then, we discuss our methodology. We then report our results and analysis. Finally, we demonstrate our conclusions, contributions and future research directions.

2. Theoretical Background and Hypotheses Development

2.1. Digital Transformation and Total Factor Productivity

Referring to resource-based theory [18,19], a firm owning heterogeneous resources, having the characteristics of value, rarity, imitability and substitutability, is more likely to obtain sustained competitive advantages and excellent performance. Consistent with prior studies [30,31], we regard digital technology as a vital heterogeneous resource contributing to sustainable high efficiency and effectiveness. Furthermore, the dynamic capability theory extends the static resource-based theory by focusing on the purposeful modification of the resource base to adapt to the external environment and maintain a firm’s superior performance [32]. Based on dynamic capability theory [20,21], upgraded digital application can be regarded as a kind of dynamic capability (capability to integrate, build and reconfigure resources to respond to the changes of external environment) that can improve a firm’s efficiency and outcomes. Thereby, both of the two dimensions of digital transformation seem to enhance a firm’s total factor productivity, which refers to the contribution of technological progress to economic growth except for factor inputs [33]. In other words, total factor productivity is the part which cannot be explained by the input growth of factors (capital and labor) [33].
To be more specific, first, digital transformation can reduce enterprise costs (major inputs) from all aspects, such as reducing labor costs, maintenance costs, time costs, production costs, service costs, etc., thereby improving a firm’s total factor productivity. With the introduction of digital technology (e.g., artificial intelligence) and digital applications (e.g., industrial internet and machine learning), a firm can dramatically reduce its reliance on human resources and subsequently reduce its labor costs [34,35]. With the integration of digital technology into traditional production tools and machines, a firm can utilize intelligent programs to monitor the entire production and service process in real time, which efficiently and effectively forecasts hidden dangers of machines and reduces maintenance costs and time costs [36]. Digital technology and digital applications alleviate information asymmetry and accelerate the flow of resources within enterprises, which improves a firm’s resource utilization efficiency and reduces its tracking and production costs [37]. In addition, a firm can take advantage of digital technology (e.g., big data, data mining and intelligent data analysis) and digital application (e.g., mobile connection and smart wear) to accurately capture consumer preference information, deliver goods and services with precision and provide customers with satisfactory after-sales service in a timely manner [38], which reduces the possibility of complaints and refunds and cuts down service costs. Consequently, digital transformation can save labor costs, maintenance costs, time costs, production costs, service costs, etc., and thus facilitate corporate total factor productivity.
Second, digital transformation can increase a firm’s technology progress outcomes (major outputs) and thus improve enterprise total factor productivity. With the introduction of digital technology (e.g., data mining) and digital applications (e.g., smart wear and smart marketing), a firm can assemble, integrate and deploy various data in real time [39]. This information contributes to breakthroughs in the current business model and creates new value for customers [40]. In addition, digital transformation is the fundamental driving force for industrial structure upgrading [41], which is widely recognized as a significant promoter of total factor productivity [42,43]. With the introduction of data as a factor of production, having the innate characteristics of high technology, high growth and platform-based communication [41], a firm can efficiently detect weak links, redundant or costly processes and missing tasks which are relevant to output in real time [26]. Finally, digital transformation expands the innovation boundary of enterprises. Digital technologies (e.g., enhanced reality and virtual reality) and digital applications (e.g., C2B, C2C and O2O) enable customers, suppliers, partners and even competitors to be deeply involved in organizational innovation processes, which enhances R&D and outcomes of new products and services [28]. Consequently, digital transformation can enhance technological progress outcomes and then facilitate corporate total factor productivity.
Thereby, the following hypotheses were proposed:
Hypothesis 1a (H1a). 
Digital technology positively affects total factor productivity.
Hypothesis 1b (H1b). 
Digital application positively affects total factor productivity.

2.2. Digital Transformation, Supplier Concentration and Total Factor Productivity

2.2.1. Digital Transformation and Supplier Concentration

Porter’s industrial competitive theory pointed out that suppliers’ bargaining power is a significant factor that influences a firm’s competitive position [22]. When only one or very few suppliers exist in the supplier market, the bargaining power of a firm will decline due to the inequality in supply-and-demand conditions. In this case, the costs of changing to a new supplier are high, and the suppliers have a cartel and exercise monopoly in pricing decisions. According to Mohapatra [44], with digital technology and applications, outsourcing orders can be automated, supplier selection can be faster, and inventory management can be automated. These features can help firms to reduce lead time for procurement and build flexible supplier–customer relationships with different vendors, which will reduce the degree of supplier concentration.
Specifically, digital technology and digital application contribute to quicker communications and truncated negotiation times between a firm and its suppliers [45]. This subsequently prompts the firm’s resilient supplier selection [46]. Due to the convenience of procuring various resources from multi-channels, the costs of changing to a new supplier will dramatically go down, and associated supplier concentration will probably decline. Furthermore, thanks to the development of digital technology and applications, the cost and speed of transmitting consolidated information simultaneously to different persons located at different geographical locations has become cheaper and faster [44]. The rapid flow of information results in more transparent outsourcing prices, which mitigates the risks caused by information imbalances and price monopolies. Consequently, the supplier concentration degree will decline. Accordingly, we propose the following hypotheses:
Hypothesis 2a (H2a). 
Digital technology negatively affects supplier concentration.
Hypothesis 2b (H2b). 
Digital application negatively affects supplier concentration.

2.2.2. Supplier Concentration and Total Factor Productivity

According to industrial competitive theory [22], with the increasing of supplier concentration, a firm’s bargaining power will be weakened when trading with its suppliers. High outsourcing prices directly raise input costs and adversely affect corporate total factor productivity. Apart from this, a concentrated supplier base increases corporate operating risks, for instance, the danger from supply shocks including sudden breakdowns or shortages of material and service supplies, unfavorable revisions in contract terms and difficulties in changing suppliers [47]. In such conditions, a firm will host more cash to deal with the potential risks, which will increase the firm’s cash-hosting costs [47]. To sum up, higher supplier concentration will raise a firm’s inputs and consequently reduce its total factor productivity.
On the other hand, a more concentrated supplier base probably impedes a firm’s technology outcomes, thus adversely affecting its total factor productivity. Specifically, when a firm mainly relies on resources and technology from one or a few regular suppliers, the firm will lose other opportunities to access and introduce new resources and technologies which are closely associated with innovation outcomes. Long-term fixed cooperation makes a firm and its suppliers keep stereotypical work norms, processes and methods that are, to some extent, not conducive to the innovative output of technology [48]. In addition, as mentioned above, when a firm hosts more cash to deal with operating risks caused by high supplier concentration, the firm will reduce its uncertain expenditures such as technological innovation investments; hence, technology outcomes will go down as well. Thereby, we propose the following hypothesis:
Hypothesis 3 (H3). 
Supplier concentration negatively affects total factor productivity.

2.2.3. The Mediating Effects of Supplier Concentration

Based on above analyses, quicker communications and truncated negotiation times brought on by digital technology and associated applications can increase supplier selection flexibility and decrease supplier concentration degree [45,46]. Furthermore, a less concentrated supplier base can improve a firm’s bargaining power, reduce its procurement costs and cash holding costs and consequently improve the firm’s total factor productivity [47]. Additionally, digital transformation makes it possible to transmit consolidated information simultaneously to different persons located at different geographical locations [44], which contributes to the transparency of outsourcing prices and channels. Under such circumstances, a less concentrated supplier base can help a firm to access and introduce more new resource portfolios, technologies, work norms and methods, improving its technology innovation outcomes and subsequently its total factor productivity [48]. Thereby, the following hypotheses are proposed:
Hypothesis 4a (H4a). 
Digital technology can promote total factor productivity through the decreasing of supplier concentration.
Hypothesis 4b (H4b). 
Digital application can promote total factor productivity through the decreasing of supplier concentration.

2.3. Digital Transformation, Customer Concentration and Total Factor Productivity

2.3.1. Digital Transformation and Customer Concentration

Transaction cost economics theory [23,24] points out that transaction cost depends on three dimensions: asset specificity, uncertainty and frequency of transaction. Asset specificity refers to the ability of an asset to be reutilized for alternative purposes by alternative users without reducing its value [23]. When a firm owns more specific assets, the firm probably prefers to control the frequency of transaction and reduce uncertainty so as to reduce transaction costs and improve transaction efficiency [24]. Based on transaction cost economics theory, with engagement in digital transformation, more digital technology and digital applications, which can be recognized as specific assets, are constantly introduced to firms. In this case, a firm is more likely to arrange various activities within its own boundaries, maximize its transaction efficiency with its customers and prefer to build high-quality major customer relationships. The degree of customer concentration will subsequently increase.
Generally, a firm and its customers are concerned with five performance factors: cost, quality, speed, dependability and flexibility [49]. Thanks to digital transformation, a firm has access to more effective communication and convenient logistics when trading with its customers [50], which guarantees the operations time, dependability and flexibility. In addition, digital technology and application improve a firm’s capability in accurate calculation and error detection, maintaining operations’ quality [51]. It has also been identified that digital economic activity can result in the reduction of operations costs, such as search costs, replication costs, transportation costs, tracking costs and verification costs [37]. Digital transformation contributes to faster delivery, more reliable delivery, higher flexibility, lower costs and higher quality of operations, which results in higher customer satisfaction and customer concentration. In addition, according to [52], digital transformation is a smart, valuable, efficient driver to create new business value for companies. Customized and enriched customer experiences brought about by digital technology and applications improve customer loyalty [53], which further increases customer concentration. Therefore, we propose the following hypotheses:
Hypothesis 5a (H5a). 
Digital technology positively affects customer concentration.
Hypothesis 5b (H5b). 
Digital application positively affects customer concentration.

2.3.2. Customer Concentration and Total Factor Productivity

Based on transaction cost economics theory, a firm with a more concentrated customer base can reduce its transaction costs and improve its transaction efficiency when trading with its customers. Customer concentration can reduce a firm’s frequency of transaction so as to save on the associated switching costs and uncertainty costs [54]. In addition, according to Huang et al. [55] and Wang and Mao [56], a firm with higher customer concentration is more likely to engage in tax avoidance and decrease its tax costs. Consequently, the firm can improve its total factor productivity due to the decrease of transaction costs, switching costs and tax costs.
On the other hand, a more concentrated customer base can help a firm to build good supplier–customer relationships with the firm’s major customers, which is helpful to improving technological innovation outputs and subsequently the total factor productivity. Specifically, a more concentrated customer base means that both a firm and its customers become more interdependent. In order to enhance integrated supply chain advantages, they prefer to collaborate to make investments, acquire proprietary assets and focus on R&D to create new value [57]. In addition, according to Peng et al. [58], customer concentration is positively related to informal financing, which, to some extent, provides financial support for a firm’s technology innovation activities. Furthermore, interdependence between companies and their customers enables both sides to engage in information sharing, particularly regarding customer demand, knowledge sharing and technology sharing, which stimulates innovative outcomes [59]. Accordingly, the following hypothesis is proposed:
Hypothesis 6 (H6). 
Customer concentration positively affects total factor productivity.

2.3.3. The Mediating Effects of Customer Concentration

Based on transaction cost economics theory and the above analyses, faster delivery, more reliable delivery, higher flexibility, lower costs and higher quality of operations brought about by digital technologies and digital applications can result in higher customer satisfaction and customer concentration. With the increase of customer concentration, the frequency of transaction and uncertainty will decline, and a firm can improve its total factor productivity due to the decrease of transaction costs, switching costs and tax costs. In addition, according to Buyukozkan and Gocer [52], digital transformation is a smart, valuable, efficient driver of creating new business value for companies. Customized and enriched customer experiences brought about by digital technology and applications improve customer loyalty [53], which further increases customer concentration. In this case, a firm and its major customers will collaborate to make investments and expand R&D to maintain their integrated supply chain advantages and subsequently improve the firm’s total factor productivity [57]. Thus, we propose the following hypotheses:
Hypothesis 7a (H7a). 
Digital technology can promote total factor productivity through the increasing of customer concentration.
Hypothesis 7b (H7b). 
Digital application can promote total factor productivity through the increasing of customer concentration.

3. Methodology

3.1. Sample Selection and Data Sources

To test the hypotheses, we collected a sample from Chinese A-share listed companies. The observation period was selected as 2007 to 2020. The sample selection processes was conducted as follows. First, we eliminated the ST, *ST (special treatment) and delisted firms. Second, we eliminated the financial firms for the reason that financial firms were not really creating new value and would probably lead to erroneous results. Third, we deleted the observations with missing items and abnormal values. As a result, we obtained 21,799 firm-year observations.
Regarding data collection, first, we obtained the annual reports of listed firms from the CNINFO website, which is the specialized information disclosure website of listed companies designated by the China Securities Regulatory Commission. We calculated the word frequency of the degree of firm digital transformation on a basis of the text content of annual reports. Second, the data associated with total factor productivity, customer concentration, supplier concentration and firm-level characteristics were collected from the database of China Stock Market and Accounting Research (CSMAR). Finally, we used the STATA 16.0 software to conduct empirical analyses.

3.2. Variable Measurements

3.2.1. Measurements of Digital Transformation

In alignment with extant studies [29,60], we measured the intensity of digital transformation through word frequency statistics from the annual reports. Specifically, first, we collected all annual reports of China’s A-share listed companies from the CNINFO website through the crawler function of the Python programming language, and then extracted all text contents through the Java PDF Box, thus forming the data pool for text analysis. Second, referring to the lexicons of digital transformation demonstrated by Wu et al. [29], we identified the keywords for text analysis (Figure 2). The paper of Wu et al. [29] was quoted by one of the most authoritative journals in China. The keywords were originally edited in Chinese. The lexicons of digital transformation are divided into two parts: the lexicons of digital technology and the lexicons of digital application. Third, we used the Python programming language to search, match and count the word frequency of all the keywords from the data pool. To ensure the scientific validity and reliability of the research, word frequency statistics in Chinese language were used for Chinese companies. After completing the word frequency analysis, we translated it into English language. In order to measure the intensity of digital transformation, we sum up the frequencies of all the keywords. Finally, considering that the explanatory variables are absolute numbers, the while mediating variables and explained variables are relative numbers, we use the logarithm of the explanatory variables for unlimited toughening treatment.

3.2.2. Measurements of Total Factor Productivity

Referring to Qian et al. [61], the production function of a firm is set as follows (Formula (1)):
Y i , t = A i , t K i , t α L i , t β M i , t γ
where Yi,t represents the output of enterprise j, and Ki,t, Li,t and Mi,t represent the amounts of capital, labor and intermediate inputs, respectively. Ai,t represents the total factor productivity. On this basis, taking the logarithm of both sides of Equation (1) and considering the factors of random interference, Equation (2) can be obtained as follows (Formula (2)):
y i , t = a i , t + α k i , t + β l i , t + γ m i , t + ε i , t
where yi,t, ai,t, ki,t, li,t and mi,t are the logarithmic forms of Yi,t, Ai,t, Ki,t, Li,t and Mi,t, respectively, and εi,t is the random disturbance term. Considering that ai,t cannot be directly observed, and ordinary least squares (OLS) estimation might cause deviations, we use the semi-parametric method proposed by Levinsohn and Petrin [62] to estimate the total factor productivity. In addition, according to Chen et al. [63], an enterprise’s main business income is used as the output index; its net fixed assets, employee wages, cash for purchasing goods and paying services constitute the input index; and the corresponding price index is used for adjustment to avoid estimation bias.

3.2.3. Measurement of Customer Concentration and Supplier Concentration

This paper utilized the purchase amount from the top five suppliers divided by the total purchase amount of an enterprise to measure supplier concentration. This is consistent with the prior study [47], and the ratio is calculated as (Formula (3)):
s u p c o i , t = ∑ i = 1 i = 5 p u r c h a s e s i , t t o t a l p u r c h a s e s t
where purchasesi,t denotes the purchases from supplier i in year t, and totalpurchasesi,t denotes the firm’s total sales in year t.
Similarly, referring to Dhaliwal et al. [64] and Campello and Gao [65], this paper used the sales of the companies’ top five customers divided by the total sales to measure customer concentration. The ratio is calculated as (Formula (4)):
c u s c o i , t = ∑ i = 1 i = 5 s a l e s i , t t o t a l s a l e s t
where salesi,t denotes the sales to customer i in year t, and totalsalesi,t denotes the firm’s total sales in year t.
All the measurements are presented in Table 1.

3.3. Model Construction

3.3.1. Models of Main Effects

The construction of a structural equation model for multiple mediation analysis can not only deal with explicit variables and latent variables, but also analyze the relationship between multiple explanatory variables, multiple explained variables and multiple mediation variables [66]. Referring to the relationships between explanatory variables and explained variables, this paper constructs the following path model of the main effects (Formula (5)):
t o t f p i , t = c 1 d i g t e i , t + c 2 d i g a p i , t + ε i , t
In Formula (5), if the path coefficients of c1 and c2 are significantly positive, H1a and H1b will be verified, respectively.

3.3.2. Models of Mediating Effects

According to the relationships among explanatory variables, mediating variables and explained variables, this paper constructs the following path models of mediating effects (Formula (6)):
{ s u p c o i , t = a 1 d i g t e i , t + a 3 d i g a p i , t + ε i , t c u s c o i , t = a 2 d i g t e i , t + a 4 d i g a p i , t + ε i , t t o t f p i , t = b 1 s u p c o i , t + b 2 c u s c o i , t + c ′ 1 d i g t e i , t + c ′ 2 d i g a p i , t + ε i , t
In Formula (6), if the path coefficients of a1 and a3 are significantly negative, H2a and H2b will be verified, respectively. If the path coefficient of b1 is significantly negative, H3 will be supported. If the mediating path coefficients of a1 × b1 (digte→supco→totfp) and a3 × b1 (digap→supco→totfp) are significantly positive, H4a and H4b will be verified, respectively. Moreover, if the path coefficients of a2 and a4 are significantly positive, H5a and H5b will respectively be verified in sequence. If the path coefficient of b2 is significantly positive, H6 will be supported. If the mediating path coefficients of a2 × b2 (digte→cosco→totfp) and a4 × b2 (digap→cosco→totfp) are significantly positive, H7a and H7b will respectively be verified in turn.

4. Results

4.1. Results of Descriptive Statistics

Table 2 illustrates the results of the descriptive statistics. The mean values of digital technology and digital application were 0.889 and 0.948, respectively, both of the minimum values of the two variables were 0, and the maximum values of the two variables were 6.140 and 6.033, respectively. This indicated that the digital technology and digital application of different firms did not vary dramatically. Similarly, the mean value of total factor productivity was 9.067, and the minimum value and maximum value of total factor productivity were 5.568 and 13.530, respectively, indicating that the total factor productivity of different firms did not vary dramatically either. Nevertheless, the mean values of supplier concentration and customer concentration were 34.100% and 30.310%, the minimum values of the two variables were 0.330% and 0.010%, and the maximum values of the two variables were 99.860% and 99.990%, respectively. This meant that the customer concentration and supplier concentration of different firms varied dramatically.

4.2. Results of Correlation Analysis

Table 3 illustrates the results of the Pearson pairwise product-moment correlation matrix among the variables. The results demonstrated that there were significant correlations among the major variables (Pearson coefficients ranging from −0.199 to 0.594). In addition, the sample data were free from multi-collinearity (all Pearson coefficients less than 0.6), a situation which is a fundamental prerequisite of structural function model analysis.

4.3. Structural Equation Model Results of the Main Effects

Figure 3 and Table 4 illustrate the structural equation model results of the main effects. The likelihood ratio and RMSEA were 0.000 and 0.041, respectively (less than 0.05), the SRMR value was 0.077 (less than 0.08), and CFI and TLT were 0.891 and 0.893, respectively (close to 0.9). This indicated acceptable goodness of fit of the main effect model [67]. Furthermore, the path coefficient of digital technology on total factor productivity was −0.012, NOT significant. Hypothesis 1a was not supported. In addition, the path coefficient of digital application on total factor productivity was 0.164, significant at 1% level. Hypothesis 1b was supported.

4.4. Structural Equation Model Results of the Mediating Effects

4.4.1. Results of the Mediating Effects

Figure 4 and Table 5 illustrate the structural equation model results of the mediating effects. The likelihood ratio and RMSEA were 0.000 and 0.026, respectively (smaller than 0.05), while the SRMR value was 0.07 (smaller than 0.08). CFI and TLT were 0.945 and 0.934, respectively (close to 1). These results indicated good goodness of fit of the mediating effect model [67].
The path coefficient of digital technology on supplier concentration and the path coefficient of digital application on supplier concentration were −0.031 and −0.077, respectively, passing the test at the 1% significance level. Hypothesis 2a and Hypothesis 2b were verified. In addition, the path coefficient of supplier concentration on total factor productivity was −0.117, significant at 1% level. Hypothesis 3 was verified. Based on the results of indirect effects of supplier concentration in Table 6, the indirect path coefficients of digte→supco→totfp and digap→supco→totfp were 0.004 and 0.009, respectively, significant at the 1% level. Hypothesis 4a and Hypothesis 4b were supported, which indicated that both digital technology and digital applications can enhance total factor productivity through the decreasing of supplier concentration.
In addition, the path coefficient of digital technology on customer concentration and the path coefficient of digital application on customer concentration were 0.049 and −0.137, respectively, passing the test at the 1% significance level. Hypothesis 5a was verified. However, regarding to Hypothesis 5b, the results indicated that digital application did not increase but rather decreased customer concentration. The reason for the opposite result might be that digital applications (e.g., mobile online, e-commerce, digital marketing, etc.) allowed the firms to rapidly provide their products and services to a wider range of customers via multi-channels. Compared with the loyalty brought about by regular major customers, the benefits brought on by new and potential customers were more obvious. Hence, Hypothesis 5b was partially verified. Furthermore, the path coefficient of customer concentration on total factor productivity was −0.155, significant at 1% level. Customer concentration did not increase but rather decreased total factor productivity. The reason for the opposite result might be that customer concentration increases operating risks and risk management costs. Meanwhile, firms might reduce uncertain technology investments and subsequently adversely affect their technological progress outcomes. Consequently, compared with the benefits brought about by customer concentration, the costs and uncertainty brought about by customer concentration were more significant. Hypothesis 6 was partially verified. Furthermore, based on the results of indirect effects of customer concentration in Table 5, the indirect path coefficient of digte→cusco→totfp was −0.008, significant at 1% level. Hypothesis 7a was partially supported, which indicated that digital technology may adversely affect total factor productivity through the increase of customer concentration. The indirect path coefficient of digap→cusco→totfp was 0.022, significant at 1% level. Hypothesis 4a was partially supported, indicating that digital application can enhance total factor productivity through the decrease of customer concentration.
Additionally, the direct path coefficient of digital technology on total factor productivity was −0.008 and was not significant. This indicated that customer concentration and supplier concentration had complete mediating effects on the pathway of digital technology affecting total factor productivity. The direct path coefficient of digital applications on total factor productivity was 0.135, significant at the 1% level. Customer concentration and supplier concentration had partial mediating effects on the relationship between digital applications and total factor productivity.

4.4.2. Results of Difference Tests of Mediating Path Coefficients

Table 7 illustrates the results of difference tests of mediating path coefficients. In the above analyses, it has been verified that digital technology probably adversely affects corporate total factor productivity through the increasing of customer concentration, while positively affect enterprise total factor productivity through the decreasing of supplier concentration. Furthermore, a coefficient difference test was conducted. The difference coefficient of a1 × b1 − a3 × b2 was 0.004, significant at 5% level. This indicated that compared with the promoting effects brought about by the decreasing of supplier concentration, the dampening effects brought about by the increasing of customer concentration were larger within the path of digital technology affecting total factor productivity. In addition, we also verified that digital applications can positively affect corporate total factor productivity through the increasing of customer concentration and the increasing of supplier concentration. Furthermore, the difference coefficient of a3 × b1 − a4 × b2 was 0.012, significant at 1% level. This pointed out that compared with the promoting effects from the increasing of supplier concentration, the promoting effects from the increasing of customer concentration were larger within the path of digital application affecting total factor productivity.

5. Discussion

5.1. Discussion of the Empirical Results

No organization or sector is immune to the influences of digital transformation [68]. Especially, in the past 3 years, the COVID-19 pandemic has brought disruptive impacts on economy, society and life, which, to some extent, further accelerated the development of corporate digital transformation [69]. In this study, we explored how digital transformation affected enterprise total factor productivity from the perspective of supply chain management. By analyzing the two dimensions of digital transformation, we found that: (1) Digital technology has no direct effects on enterprise total factor productivity (with a path coefficient of −0.012, not significant), while digital application is directly positively related to total factor productivity (with a path coefficient of 0.164, significant at the 1% level). It is difficult for digital technology to directly improve corporate total factor productivity for the reason that the digital technology is more closely related to fundamental digital infrastructure (e.g., big data technology and cloud computing technology). In this early stage, the inputs of digital technology and infrastructure are usually expensive and time-consuming, and it is difficult to directly transfer them to corporate main business income in the short term. In comparison, digital applications (e.g., smart agriculture and smart medical care) can be directly applied in corporate operations, bringing business incomes and improving corporate total factor productivity in the short term. (2) Both digital technology and digital application can indirectly facilitate enterprise total factor productivity through the decreasing of supplier concentration (with indirect path coefficients of 0.004 and 0.009, respectively, significant at the 1% level). Accordingly, the convenient information and communication brought about by digital technology and applications can increase supplier selection flexibility, which, in turn, improves a firm’s bargaining power with its suppliers, reduces its procurement and cash holding costs and consequently improves the firm’s total factor productivity through cost saving. (3) Differently from theoretical hypothesis proposed, digital technology adversely affects enterprise total factor productivity through the increasing of customer concentration (with an indirect path coefficient of −0.008, significant at the 1% level); meanwhile, digital application may positively affect enterprise total factor productivity, but does so through the decreasing of customer concentration (with an indirect path coefficient of 0.021, significant at the 1% level). Digital technology will increase customer concentration and further adversely affects corporate total factor productivity for the reason that customer concentration might increase operating risks and risk management costs. Firms might reduce uncertain technology investments and subsequently adversely affect their technological progress outcomes. Compared with the benefits brought about by customer concentration, the costs and uncertainty brought about by customer concentration were more significant. Digital application will increase supplier concentration and further improve corporate total factor productivity because digital application allows firms to rapidly provide their products and services to a wider range of customers via multi-channels. Compared with the loyalty of regular major customers, the benefits brought about by new and potential customers were more obvious.

5.2. Theoretical Implications

Our research results contribute to the extent literature in the following aspects. First, to our knowledge, our study is one of the first to decouple digital transformation into digital technology and digital application and propose a new conceptual framework highlighting the link of digital technology and digital application with total factor productivity. The extant literature that focused on the relationship between digital transformation and total factor productivity failed to distinguish the different roles played by digital technology and digital application in total factor productivity (e.g., Pan et al. [7], Zeng and Lei [8] and Ren et al. [70]). The verification of our conceptual framework provides empirical evidence for the relationship of digital technology and digital application, the two dimensions of digital transformation, with total factor productivity, therefore filling a gap in the extant literature. Second, our study is one of the first to explore the relationship between the dimensions of digital transformation and enterprise total factor productivity from the perspective of supply chain management. The extant literature proposed that digital transformation can improve total factor productivity through the mediator of technical cooperation [71]. We identified the impacts of digital transformation on enterprise total factor productivity mediated by the multiple mediators of supplier concentration and customer concentration, which filled the gap of mediating pathway research in the existing literature. Third, we further conducted difference tests of the two mediating pathways of supplier concentration and customer concentration, expanding the depth and breadth of the existing research.

5.3. Practical Implications

Our research also provides some practical implications for corporate leaders and managers. First, in the digital era, legacy firms should focus on digital transformation. Especially, members of top management are advised to introduce digital application into firms, which is a significant driver of the improvement of total factor productivity. Second, when a firm engages in digital transformation, the top management is recommended to flexibly select suppliers. With the decrease of supplier concentration, the firm can gain more opportunities to improve total factor productivity. Third, when a firm engages in digital transformation, the top management should be cautious with the introduction of digital technology for the reason that digital technology probably increases the firm’s customer concentration and then adversely affects the firm’s total factor productivity. In addition, compared with the promoting effects brought about by the decreasing of supplier concentration, the dampening effects brought about by the increasing of customer concentration were larger within the path of digital technology affecting total factor productivity.

5.4. Limitations and Directions for Future Research

Although this paper provides some theorical and practical implications to scholars and managers, it has some limitations. First, we use a large sample covering all the Chinese A-listed companies over the period from 2007 to 2020. When we generalize our findings to companies in other national contexts, we need to consider whether the findings are still tenable and applicable. Second, the intensity of digital transformation is measured by word frequency statistics from annual reports of listed companies. We use more than 80 keywords associated with digital transformation. Nevertheless, because of the rapid development of digital economy, we cannot cover all the keywords related to digital transformation in real time. These limitations help us to determine future research directions. We can collect more data in other countries to further investigate the robustness of our findings and expand our keyword selection scope related to digital transformation.

6. Conclusions

Using a sample of 21,799 firm-year observations of Chinese listed companies over the period from 2007 to 2020, our study empirically investigated whether digital transformation could enhance enterprise total factor productivity through the decreasing of supplier concentration and increasing of customer concentration. The research reveals the following: First, digital technology has no significant effects on enterprise total factor productivity, while digital application can directly enhance enterprise total factor productivity. Second, both digital technology and digital application can indirectly boost enterprise total factor productivity through the decreasing of supplier concentration. Third, digital technology adversely affects enterprise total factor productivity through the increasing of customer concentration; however, digital application may positively affect enterprise total factor productivity through the decreasing of customer concentration. In theory, our study is one of the first to explore the mediating effect mechanism of the relationship between digital transformation and enterprise total factor productivity from the perspective of supply chain management. In practice, based on our findings, companies are suggested to vigorously promote digital transformation, especially by introducing digital application into firms to improve total factor productivity. A firm engaging in digital transformation is recommended to flexibly select suppliers to improve total factor productivity. When a firm engages in digital transformation, the top management should be cautious with the introduction of digital technology because it probably increases the firm’s customer concentration and then adversely affects the firm’s total factor productivity.

Author Contributions

Conceptualization, H.Z.; methodology, H.Z. and Q.Z.; software, H.Z. and Q.Z.; validation, H.Z. and Q.Z.; formal analysis, H.Z. and Q.Z.; investigation, H.Z. and Q.Z.; resources, H.Z.; data curation, H.Z. and Q.Z.; writing—original draft preparation, H.Z.; writing—review and editing, Q.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Education Department of Liaoning Province, China, grant number LJKMR20220424.

Data Availability Statement

Data are available from authors upon reasonable request.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Majchrzak, A.; Markus, M.L.; Wareham, J. Designing for digital transformation: Lessons for information systems research from the study of ICT and societal challenges. MIS Quart. 2016, 40, 267–277. [Google Scholar] [CrossRef]
  2. Vial, G. Understanding digital transformation: A review and a research agenda. J. Strategic Inf. Syst. 2019, 28, 118–144. [Google Scholar] [CrossRef]
  3. Frank, A.G.; Mendes, G.H.S.; Ayala, N.F.; Ghezzi, A. Servitization and Industry 4.0 convergence in the digital transformation of product firms: A business model innovation perspective. Technol. Forecast. Soc. 2019, 141, 341–351. [Google Scholar] [CrossRef]
  4. Llopis-Albert, C.; Rubio, F.; Valero, F. Impact of digital transformation on the automotive industry. Technol. Forecast. Soc. 2021, 162, 120343. [Google Scholar] [CrossRef] [PubMed]
  5. Guenzi, P.; Habel, J. Mastering the Digital Transformation of Sales. Calif. Manage. Rev. 2020, 62, 57–85. [Google Scholar] [CrossRef]
  6. Henderson, D. Demand-side broadband policy in the context of digital transformation: An examination of SME digital advisory policies in Wales. Telecommun. Policy 2020, 44, 102024. [Google Scholar] [CrossRef]
  7. Pan, W.R.; Xie, T.; Wang, Z.W.; Ma, L.S. Digital economy: An innovation driver for total factor productivity. J. Bus. Res. 2021, 139, 303–311. [Google Scholar] [CrossRef]
  8. Zeng, G.Y.; Lei, L.X. Digital transformation and corporate total factor productivity: Empirical evidence based on listed enterprises. Discrete. Dyn. Nat. Soc. 2021, 2021, 9155861. [Google Scholar] [CrossRef]
  9. Feng, C.; Huang, J.B.; Wang, M. Analysis of green total-factor productivity in China’s regional metal industry: A meta-frontier approach. Resour. Policy 2018, 58, 219–229. [Google Scholar] [CrossRef]
  10. Amri, F.; Ben Zaied, Y.; Ben Lahouel, B. ICT, total factor productivity, and carbon dioxide emissions in Tunisia. Technol. Forecast. Soc. 2019, 146, 212–217. [Google Scholar] [CrossRef]
  11. Jin, W.; Zhang, H.Q.; Liu, S.S.; Zhang, H.B. Technological innovation, environmental regulation, and green total factor efficiency of industrial water resources. J. Clean. Prod. 2019, 211, 61–69. [Google Scholar] [CrossRef]
  12. Du, K.R.; Li, J.L. Towards a green world: How do green technology innovations affect total-factor carbon productivity. Energ. Policy 2019, 131, 240–250. [Google Scholar] [CrossRef]
  13. Wang, H.R.; Cui, H.R.; Zhao, Q.Z. Effect of green technology innovation on green total factor productivity in China: Evidence from spatial durbin model analysis. J. Clean. Prod. 2021, 288, 125624. [Google Scholar] [CrossRef]
  14. Song, M.L.; Peng, L.C.; Shang, Y.P.; Zhao, X. Green technology progress and total factor productivity of resource-based enterprises: A perspective of technical compensation of environmental regulation. Technol. Forecast. Soc. 2022, 174, 121276. [Google Scholar] [CrossRef]
  15. Gaglio, C.; Kraemer-Mbula, E.; Lorenz, E. The effects of digital transformation on innovation and productivity: Firm-level evidence of South African manufacturing micro and small enterprises. Technol. Forecast. Soc. 2022, 182, 121785. [Google Scholar] [CrossRef]
  16. Ferreira, J.J.M.; Fernandes, C.I.; Ferreira, F.A.F. To be or not to be digital, that is the question: Firm innovation and performance. J. Bus. Res. 2019, 101, 583–590. [Google Scholar] [CrossRef]
  17. Nambisan, S.; Wright, M.; Feldman, M. The digital transformation of innovation and entrepreneurship: Progress, challenges and key themes. Res. Policy 2019, 48, 103773. [Google Scholar] [CrossRef]
  18. Barney, J.B. The resource-based theory of the firm. Organ. Sci. 1996, 7, 469. [Google Scholar] [CrossRef]
  19. Barney, J.B. Strategic management: From informed conversation to academic discipline. Acad. Manag. Perspect. 2002, 16, 53–57. [Google Scholar] [CrossRef]
  20. Teece, D.J.; Pisano, G.; Shuen, A. Dynamic capabilities and strategic management. Strategic Manag. J. 1997, 18, 509–533. [Google Scholar] [CrossRef]
  21. Teece, D.; Peteraf, M.; Leih, S. Dynamic capabilities and organizational agility: Risk, uncertainty and entrepreneurial management in the innovation economy. Calif. Manag. Rev. 2016, 58, 13–35. [Google Scholar] [CrossRef] [Green Version]
  22. Porter, M.E. How competitive forces shape strategy. Harvard Bus. Rev. 1979, 57, 137. [Google Scholar]
  23. Williamson, O.E. Markets and Hierarchies: Analysis and Antitrust Implications, 1st ed.; Free Press: New York, NY, USA, 1975. [Google Scholar]
  24. Aubert, B.A.; Rivard, S.; Patry, M. A transaction cost approach to outsourcing behavior: Some empirical evidence. Inform. Manag.-Amster. 1996, 30, 51–64. [Google Scholar] [CrossRef]
  25. Liu, D.Y.; Chen, S.W.; Chou, T.C. Resource fit in digital transformation. Manage. Decis. 2012, 49, 1728–1742. [Google Scholar] [CrossRef]
  26. Kretschmer, T.; Khashabi, P. Digital transformation and organization design: An integrated approach. Calif. Manag. Rev. 2020, 62, 86–104. [Google Scholar] [CrossRef]
  27. Jafari-Sadeghi, V.; Garcia-Perez, A.; Candelo, E.; Couturier, J. Exploring the impact of digital transformation on technology entrepreneurship and technological market expansion: The role of technology readiness, exploration and exploitation. J. Bus. Res. 2021, 124, 100–111. [Google Scholar] [CrossRef]
  28. Yang, M.Y.; Fu, M.T.; Zhang, Z.H. The adoption of digital technologies in supply chains: Drivers, process and impact. Technol. Forecast. Soc. 2021, 169, 120795. [Google Scholar] [CrossRef]
  29. Wu, F.; Hu, H.Z.; Lin, H.Y.; Ren, X.Y. Enterprise digital transformation and capital market performance: Empirical evidence from stock liquidity. Manag. World 2021, 37, 130–144+10. [Google Scholar]
  30. Elia, S.; Giuffrida, M.; Mariani, M.M.; Bresciani, S. Resources and digital export: An RBV perspective on the role of digital technologies and capabilities in cross-border e-commerce. J. Bus. Res. 2021, 132, 158–169. [Google Scholar] [CrossRef]
  31. Giustiziero, G.; Kretschmer, T.; Somaya, D.; Wu, B. Hyperspecialization and hyperscaling: A resource-based theory of the digital firm. Strateg. Manag. J. 2022; early access. [Google Scholar] [CrossRef]
  32. Schilke, O.; Hu, S.; Helfat, C.E. Quo Vadis, Dynamic Capabilities? A Content Analytic Review of the Current State of Knowledge and Recommendations for Future Research. Acad. Manag. Ann. 2018, 12, 390–439. [Google Scholar] [CrossRef] [Green Version]
  33. Sun, J.R.; Zhang, J.; Zhang, J.H.; Ma, J.H.; Zhang, Y.L. Total factor productivity assessment of tourism industry: Evidence from China. Asia. Pac. J. Tour. Res. 2015, 20, 280–294. [Google Scholar] [CrossRef]
  34. Acemoglu, D.; Restrepo, P. The race between man and machine: Implications of technology for growth, factor shares, and employment. Am. Econ. Rev. 2018, 108, 1488–1542. [Google Scholar] [CrossRef] [Green Version]
  35. Tu, W.; He, J. Can digital transformation facilitate firms’ M&A: Empirical discovery based on machine learning. Emerg. Mark. Financ. Tr. 2022, 59, 113–128. [Google Scholar]
  36. Ansari, F. Cost-based text understanding to improve maintenance knowledge intelligence in manufacturing enterprises. Comput. Ind. Eng. 2020, 141, 106319. [Google Scholar] [CrossRef]
  37. Goldfarb, A.; Tucker, C. Digital economics. J. Econ. Lit. 2019, 57, 3–43. [Google Scholar] [CrossRef] [Green Version]
  38. Schweidel, D.A.; Bart, Y.; Inman, J.J.; Stephen, A.T.; Libai, B.; Andrews, M.; Rosario, A.B.; Chae, I.; Chen, Z.; Kupor, D.; et al. How consumer digital signals are reshaping the customer journey. J. Acad. Market. Sci. 2022, 50, 1257–1276. [Google Scholar] [CrossRef]
  39. Gupta, M.; George, J.F. Toward the development of a big data analytics capability. Inform. Manag.-Amster. 2016, 53, 1049–1064. [Google Scholar] [CrossRef]
  40. Ghezzi, A.; Cavallo, A. Agile business model innovation in digital entrepreneurship: Lean startup approaches. J. Bus. Res. 2020, 110, 519–537. [Google Scholar] [CrossRef]
  41. Liu, Y.; Yang, Y.L.; Li, H.H.; Zhong, K.Y. Digital economy development, industrial structure upgrading and green total factor productivity: Empirical evidence from China’s cities. Int. J. Env. Res. Pub. He. 2022, 19, 2414. [Google Scholar] [CrossRef]
  42. Lu, X.H.; Jiang, X.; Gong, M.Q. How land transfer marketization influence on green total factor productivity from the approach of industrial structure? Evidence from China. Land Use Policy 2020, 95, 104610. [Google Scholar] [CrossRef]
  43. Yu, B.B. The impact of the internet on industrial green productivity: Evidence from China. Technol. Forecast. Soc. 2022, 177, 121527. [Google Scholar] [CrossRef]
  44. Mohapatra, S. IT and Porter’s competitive forces model and strategies, information systems theory: Explaining and predicting our digital society. Integr. Ser. Inf. Syst. 2010, 28, 265–281. [Google Scholar]
  45. Obal, M.; Lancioni, R.A. Maximizing buyer-supplier relationships in the digital era: Concept and research agenda. Ind. Market. Manag. 2013, 42, 851–854. [Google Scholar] [CrossRef]
  46. Cavalcante, I.M.; Frazzon, E.M.; Forcellini, F.A.; Ivanov, D. A supervised machine learning approach to data-driven simulation of resilient supplier selection in digital manufacturing. Int. J. Inform. Manag. 2019, 49, 86–97. [Google Scholar] [CrossRef]
  47. Zhang, X.D.; Zou, M.F.; Liu, W.M.; Zhang, Y.F. Does a firm’s supplier concentration affect its cash holding? Econ. Model. 2020, 90, 527–535. [Google Scholar] [CrossRef]
  48. Tang, X.; Rai, A. The moderating effects of supplier portfolio characteristics on the competitive performance impacts of supplier-facing process capabilities. J. Oper. Manag. 2012, 30, 85–98. [Google Scholar] [CrossRef]
  49. Slack, N.; Brandon-Jones, A.; Johnston, R. Essentials of Operations Management, 1st ed.; Prentice Hall: Upper Saddle River, NJ, USA, 2011. [Google Scholar]
  50. Cheng, Z.; Nault, B.R. Relative industry concentration and customer-driven IT spillovers. Inform. Syst. Res. 2012, 23, 340–355. [Google Scholar] [CrossRef] [Green Version]
  51. Wagner, R.; Haefner, B.; Biehler, M.; Lanza, G. Digital DNA in quality control cycles of high-precision products. CIRP Ann-Manuf. Techn. 2020, 69, 373–376. [Google Scholar] [CrossRef]
  52. Buyukozkan, G.; Gocer, F. Digital supply chain: Literature review and a proposed framework for future research. Comput. Ind. 2018, 97, 157–177. [Google Scholar] [CrossRef]
  53. Kopalle, P.K.; Kumar, V.; Subramaniam, M. How legacy firms can embrace the digital ecosystem via digital customer orientation. J. Acad. Market. Sci. 2020, 48, 114–131. [Google Scholar] [CrossRef]
  54. Ma, X.F.; Wang, W.M.; Wu, J.G.; Zhang, W.L. Corporate customer concentration and stock price crash risk. J. Bank Financ. 2020, 119, 105903. [Google Scholar] [CrossRef]
  55. Huang, H.H.; Lobo, G.J.; Wang, C.; Xie, H. Customer concentration and corporate tax avoidance. J. Bank Financ. 2016, 72, 184–200. [Google Scholar] [CrossRef]
  56. Wang, J.H.; Mao, N. Customer and tax behaviour: How customer concentration affects suppliers’ tax avoidance. Account. Forum 2021, 45, 363–388. [Google Scholar] [CrossRef]
  57. Shen, H.Y.; Xia, N.; Zhang, J. Customer-based concentration and firm innovation. Asia-Pac. J. Financ. St. 2018, 47, 248–279. [Google Scholar] [CrossRef]
  58. Peng, X.A.; Wang, X.Y.; Yan, L.N. How does customer concentration affect informal financing? Int. Rev. Econ. Financ. 2019, 63, 152–162. [Google Scholar] [CrossRef]
  59. Zhao, W.X.; Wang, C.Y.; Wan, L.; Wang, Q.; Luo, B.A. Customer concentration and exploratory innovation: The mediating effect of perceived performance-reducing threats. J Bus. Econ. Manag. 2021, 22, 940–957. [Google Scholar] [CrossRef]
  60. Zhang, J.C.; Long, J.; Schaewen, A.M.E. How does digital transformation improve organizational resilience? Findings from PLS-SEM and fsQCA. Sustainability 2021, 13, 11487. [Google Scholar] [CrossRef]
  61. Qian, X.; Kang, J.; Tang, Y. Industrial policy, efficiency of capital allocation and firm’s total factor productivity: Evidence from a natural experiment in China. China Ind. Econ. 2018, 35, 42–59. [Google Scholar]
  62. Levinsohn, J.; Petrin, A. Estimating production functions using inputs to control for un-observables. Rev. Econ. Stud. 2003, 70, 317–341. [Google Scholar] [CrossRef]
  63. Chen, Y.; Yang, S.P.; Li, Q. How does the development of digital financial inclusion affect the total factor productivity of listed companies? Evidence from China. Financ. Res. Lett. 2022, 47, 102956. [Google Scholar] [CrossRef]
  64. Dhaliwal, D.; Judd, J.S.; Serfling, M.; Shaikh, S. Customer concentration risk and the cost of equity capital. J. Account. Econ. 2016, 61, 23–48. [Google Scholar] [CrossRef]
  65. Campello, M.; Gao, J. Customer concentration and loan contract terms. J. Financ. Econ. 2017, 123, 108–136. [Google Scholar] [CrossRef]
  66. Bagozzi, R.P.; Yi, Y. Specification, evaluation, and interpretation of structural equation models. J. Acad. Market. Sci. 2012, 40, 8–34. [Google Scholar] [CrossRef]
  67. Kline, R.B. Principles and Practice of Structural Equation Modeling, 4th ed.; Guilford Press: New York, NY, USA, 2016. [Google Scholar]
  68. Hess, T.; Matt, C.; Benlian, A.; Wiesbock, F. Options for formulating a digital transformation strategy. MIS Quart. 2016, 15, 123–139. [Google Scholar]
  69. Rupeika-Apoga, R.; Petrovska, K.; Bule, L. The effect of digital orientation and digital capability on digital transformation of SMEs during the COVID-19 pandemic. J. Theor. Appl. Electron. Commer. Res. 2022, 17, 669–685. [Google Scholar] [CrossRef]
  70. Ren, Y.J.; Zhang, X.; Chen, H. The impact of new energy enterprises’ digital transformation on their total factor productivity: Empirical evidence from China. Sustainability 2022, 14, 13928. [Google Scholar] [CrossRef]
  71. Li, N.; Wang, X.H.; Wang, Z.R.; Luan, X.Y. The impact of digital transformation on corporate total factor productivity. Front. Psychol. 2022, 13, 1071986. [Google Scholar] [CrossRef]
Sustainability 15 01896 g001 550
Figure 1. (a) Digital economy development index in China (2012–2020); (b) Total factor productivity of listed digital enterprises in China (2012–2020).
Figure 1. (a) Digital economy development index in China (2012–2020); (b) Total factor productivity of listed digital enterprises in China (2012–2020).
Sustainability 15 01896 g001
Sustainability 15 01896 g002 550
Figure 2. Keywords for text analysis.
Figure 2. Keywords for text analysis.
Sustainability 15 01896 g002
Sustainability 15 01896 g003 550
Figure 3. Path diagram and empirical results of main effects. Note: *** represents that it is significant at the 1% level.
Figure 3. Path diagram and empirical results of main effects. Note: *** represents that it is significant at the 1% level.
Sustainability 15 01896 g003
Sustainability 15 01896 g004 550
Figure 4. Path diagram and empirical results of mediating effects. Note: *** represents that it is significant at the 1% level.
Figure 4. Path diagram and empirical results of mediating effects. Note: *** represents that it is significant at the 1% level.
Sustainability 15 01896 g004
Table
Table 1. Definitions of variables.
Table 1. Definitions of variables.
TypeVariableSymbolMeasurement
Explanatory
variables		Digital
technology		digteUse PYTHON to conduct text analysis of the annual reports of listed firms to form keyword frequency statistics
Digital
application		digapUse PYTHON to conduct text analysis of the annual reports of listed firms to form keyword frequency statistics
Explained
variable		Total factor
productivity		totfpCalculate efficiency of enterprises transforming various inputs (net fixed assets, employee wages, cash for purchasing goods and paying services) into outputs (main business income)
Mediating variablesSupplier
concentration		supcoPurchases from the firms’ top five suppliers divided by the total purchases
Customer
concentration		cuscoSales to the firms’ top five customers divided by the total sales
Table
Table 2. Results of descriptive statistics.
Table 2. Results of descriptive statistics.
VariablesNumberMeanStandard DeviationMinimumMaximum
digte21,7990.8891.2540.0006.140
digap21,7990.9481.1560.0006.033
totfp21,7999.0671.0955.56813.530
supco (%)21,79934.10019.6800.33099.860
cusco (%)21,79930.31021.5400.01099.990
Table
Table 3. Results of correlation analysis.
Table 3. Results of correlation analysis.
VariablesDigteDigapSupcoCuscoTotfp
digte1
digap0.594 ***1
supco−0.077 ***−0.095 ***1
cusco−0.032 ***−0.108 ***0.263 ***1
totfp0.086 ***0.157 ***−0.169 ***−0.199 ***1
Note: *** represents that it is significant at the 1% level.
Table
Table 4. Results of main effects.
Table 4. Results of main effects.
EffectsCoefficientsStandard ErrorZ Valuep Value95% CI
c1−0.012 0.008 −1.410 0.159 −0.028 0.005
c20.164 0.008 20.000 0.000 0.148 0.180
constant8.151 0.041 198.080 0.000 8.071 8.232
Table
Table 5. Results of mediating effects.
Table 5. Results of mediating effects.
EffectsCoefficientsStandard ErrorZ Valuep Value95% CI
a1−0.031 0.008 −3.670 0.000 −0.047 −0.014
a3−0.077 0.008 −9.250 0.000 −0.094 −0.061
constant1.818 0.012 152.990 0.000 1.795 1.841
a20.049 0.008 5.870 0.000 0.033 0.065
a4−0.137 0.008 −16.500 0.000 −0.153 −0.121
constant1.485 0.011 135.910 0.000 1.463 1.506
b1−0.117 0.007 −17.190 0.000 −0.130 −0.104
b2−0.155 0.007 −22.870 0.000 −0.168 −0.142
c1′ −0.008 0.008 −0.950 0.342 −0.024 0.008
c2′ 0.135 0.008 16.560 0.000 0.119 0.151
constant8.631 0.042 204.240 0.000 8.549 8.714
Table
Table 6. Results of significance test of indirect effects.
Table 6. Results of significance test of indirect effects.
EffectsCoefficientsStandard ErrorZ Valuep Value95% CI
a1 × b10.004 0.001 3.590 0.000 0.002 0.006
a3 × b10.009 0.001 8.140 0.000 0.007 0.011
a3 × b2−0.008 0.001 −5.690 0.000 −0.010 −0.005
a4 × b20.021 0.002 13.350 0.000 0.018 0.024
Table
Table 7. Results of path coefficient difference test.
Table 7. Results of path coefficient difference test.
EffectsCoefficientsStandard ErrorZ Valuep Value95% CI
a1 × b1 − a3 × b20.004 0.002 2.380 0.017 0.001 0.007
a3 × b1 − a4 × b20.012 0.002 6.060 0.000 0.008 0.016
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

Zhang, H.; Zhang, Q. How Does Digital Transformation Facilitate Enterprise Total Factor Productivity? The Multiple Mediators of Supplier Concentration and Customer Concentration. Sustainability 2023, 15, 1896. https://doi.org/10.3390/su15031896

AMA Style

Zhang H, Zhang Q. How Does Digital Transformation Facilitate Enterprise Total Factor Productivity? The Multiple Mediators of Supplier Concentration and Customer Concentration. Sustainability. 2023; 15(3):1896. https://doi.org/10.3390/su15031896

Chicago/Turabian Style

Zhang, Hua, and Qiwang Zhang. 2023. "How Does Digital Transformation Facilitate Enterprise Total Factor Productivity? The Multiple Mediators of Supplier Concentration and Customer Concentration" Sustainability 15, no. 3: 1896. https://doi.org/10.3390/su15031896

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop