Digital Technological Diversity: The Root Cause of Export Vulnerability for Enterprises in Adversity?

Abstract

There is no consensus in existing research on whether the application of digital technology diversification strengthens or weakens export resilience. This study fills this gap by exploring the influence mechanism of digital technology diversity on enterprise export resilience. Based on organizational inertia theory, dynamic capabilities perspective, and organizational learning theory, this study uses data from Chinese listed export manufacturing companies from 2019 to 2023 in order to empirically examine the relationship between digital technology diversity and enterprise export resilience. The results show that digital technology diversity significantly inhibits enterprise export resilience, supply chain finance plays a partially mediating role in this relationship, and digital interlock alleviates the inhibitory effect of digital technology diversity on supply chain finance. The findings provide guidance for the digital technology application strategy of export enterprises in adversity.

1. Introduction

Exports are an important way for enterprises to develop international markets and realize the internationalization process, which plays an important role in driving industrial upgrading and improving the socio-economic pattern [1]. However, the occurrence of Black Swan Even events, such as a trade war between the USA and China, COVID-19, and the Russian–Ukrainian conflict, have intensified the operational risk of export enterprises, resulting in many enterprises experiencing supply chain disruptions and capital chain breakage. The impact of such uncontrollable events drives the transformation of corporate strategic paradigms from profit-oriented to resilience building [2]. In the era of the digital economy, digital technology provides opportunities to establish resilience from traditional strategic paradigms to long-term sustainable development orientation.

Studies in Jordan and Italy have shown that big data analytics capabilities and cross-border e-commerce are crucial for export manufacturing companies to quickly recover exports during crises [3,4]. Specifically, digital technology applications, such as big data analysis, AI, cross-border e-commerce, and blockchain, enhance the dynamic ability of enterprises to perceive market changes, seize opportunities, and reconfigure internal and external resources [5], which facilitates exporters to quickly adjust their export strategies and access new export markets [3,4]. However, digital technology itself cannot directly create additional returns, and its value depends on whether enterprises can redeploy strategic decisions based on the capabilities of digital technology [6]. Research has shown that digital technology plays a fundamental role in promoting information exchange between enterprises and supply chain partners [7,8] and is also an important tool for implementing supply chain finance [9]. On the one hand, digital technology’s capabilities in information collection and integration enhance the visibility and data quality among supply chain participants. This, in turn, enables financial providers to comprehensively assess corporate credit risk [10], thereby improving the accessibility of supply chain finance. On the other hand, export enterprises are vulnerable to credit defects and need to ensure adequate working capital [11].

However, existing research has also noted that enterprises are often constrained by rigid digital technology architectures and complex nesting of different digital technologies, resulting in reduced information transparency and enterprise agility. For example, the lack of standardized digital technology interfaces and unified protocols makes it difficult to achieve effective vertical and horizontal integration [12], which in turn increases the complexity of information processing. Moreover, when automation processes and backbone networks need to change, the close integration of IT and business processes can easily lead to routine rigidity [13]. Despite these insights, existing research mostly focuses on the perspective of a single technology or static technological capabilities, ignoring the reality that enterprises usually require multi-technology collaborative applications in actual operations. Therefore, a key research gap exists in understanding how digital technological diversity impacts export resilience. To address this gap, this study employs supply chain finance as a mediating role to explore the impact of digital technological diversity on export resilience.

The dominant view is that inertia stems from the subject’s adherence to their cognitive framework and subjective structure [14]. Therefore, resource rigidity and routine rigidity induced by digital technological diversity depend on enterprises’ cognition. Organizational learning theory further suggests that digital interlock formed among enterprises enriches the knowledge base, perspectives, and cognition of digital technology [15,16]. This enriched cognition significantly influences the organizational inertia induced by digital technological diversity. Interlocking directorate is one of the representative ways of enterprise interlock [17]. By serving as outside directors in other digital transformation enterprises, directors could acquire tacit knowledge of digital technology, inside information on technological change, and contextual knowledge of technological applications (hereafter termed “digital interlock”) [15]. Therefore, this study further analyzes the moderation effect of digital interlock on the relationship between digital technological diversity and supply chain finance.

Based on organizational inertia theory, dynamic capability view, and organizational learning theory, this study contributes to the literature on digital technology and export resilience from the following three aspects. First, existing studies mostly focus on the positive impacts of digital technology [3,4,5]. This study considers the resource rigidity and routine rigidity generated by technology nesting and finds that digital technological diversity reduces enterprise export resilience during the crisis. This finding responds to recent calls for further investigations into the potential negative impact of digital technology on enterprise production and operational performance. Second, focusing on the relational and informational capabilities adjusted by digital technology applications, this study investigates the mediation role of supply chain finance in the relationship between digital technological diversity and export resilience, which contributes to a deeper understanding of the mediation mechanism. Finally, this study attempts to explore the moderation effect of digital interlock, providing a new perspective on how enterprises can overcome inertia and build dynamic capabilities through organizational learning and knowledge integration.

2. Theoretical Development and Hypotheses

Resilience corresponds to disruptions, and in the field of operations management, resilience is defined as the ability of an enterprise to resist disruptions and recover quickly when faced with shocks or emergency situations [18]. Consistent with existing research, this study emphasizes the understanding of enterprise export resilience by highlighting their export resistance ability during crises and their export recovery ability after crises. The establishment of enterprise export resilience is more dependent on their endowment in acquiring, allocating, and utilizing resources [19], which challenges their dynamic ability to efficiently match resource elements in response to crises. Digital technology has changed business processes, business models, and enterprise capabilities [20], especially in resource-poor emerging markets, where many enterprises regard digital technology as a key factor in developing dynamic capabilities to enhance exports [21]. For example, Zhang et al. (2024) [22] found that cross-border e-commerce platforms can collect data on customer behaviors in cross-border markets, which helps enterprises achieve customized production and promotes diversified export products. Yoon et al. (2020) [23] demonstrated that blockchain technology can shorten delivery time and transportation costs and reduce export risks. It can be seen that digital technology advantages are crucial for enterprises to adapt and continuously adjust their dynamic capabilities to cope with the volatile external environment.

The dynamic capability view emphasizes that enterprises need to possess two fundamental dynamic capabilities to gain competitive advantage: relational capability and information capability [24]. Relationship capability refers to an enterprise’s ability to collaborate and build partnerships within the ecosystem, which facilitates improved adaptability and absorptive capacity in dynamic environments [25]. Information capability refers to the ability of timely and efficient information exchange and sharing between enterprises and partners, which helps to enhance the transparency of enterprise information [26]. Supply chain finance encompasses both financial and supply chain functions, surpassing the boundaries of a single enterprise, and refers to the integration of logistics, information, and funds among various enterprises to swiftly address the financial requirements of enterprises in the supply chain [27]. Its essence lies in information sharing and efficient collaboration among different enterprises in the supply chain. Therefore, this study regards the supply chain finance activities carried out by enterprises as their relationship capabilities and information capabilities in order to study the relationship between digital technology, supply chain finance, and export resilience, which is consistent with the theoretical framework of the dynamic capability view.

However, the organizational inertia theory suggests that despite changes in the external environment, companies still tend to maintain their current strategic decision-making and operational models, manifested as resource rigidity and conventional rigidity [14]. In reality, digital technology is often deeply tied to business processes and becomes a core component of complex production and operation systems, making it difficult for enterprises to easily replace technology or adjust processes [28]. This technology–process lock-in effect makes enterprises rely on existing institutional practices to maintain efficiency, but also reduces their flexibility in responding to sudden shocks [14]. Therefore, although digital technology has the potential to enhance resilience, diversified applications may lead to technological fragmentation, which impedes firms’ dynamic capabilities in the face of environmental changes. Organizations learn by integrating previous experiences and obtaining performance feedback [29], but the path dependence of organizational learning often leads to short-sighted behavior that hinders corporate change [30]. Therefore, acquiring external knowledge is an important strategy for organizations to overcome internal inertia. Existing research emphasizes that inter-firm connections are the source of heterogeneous knowledge learning. Organizational learning occurs when firms integrate external knowledge and prior experience to correct or supplement their own shortcomings [31]. Extensive research within organizational learning theory indicates that enterprise interlock serves as an effective way for organizational knowledge acquisition and strategic adaptation. This is largely attributed to interlocking directors’ comprehensive expertise in technological applications, operational processes, and development strategies [32]. Therefore, this study suggests that directors serving in external digital transformation enterprises, namely digital interlock, can enhance their own digital technology expertise and enrich their perception of digital technology applications.

2.1. The Effect of Digital Technological Diversity on Enterprise Export Resilience

Digital technological diversity may weaken enterprise export resilience. Firstly, digital technological diversity in production is a source of conventional inertia that prevents production mode adjustment and reduces the flexibility of intelligent production in exporting. Digital technology is embedded in various aspects of enterprises’ production processes, and the high coupling between systems and components makes them highly sensitive to external environmental changes and newly introduced elements [33]. Therefore, the greater variety of digital technologies applied by a firm after an external shock can instead undermine the stability and reliability of its production operations, weakening its export defense ability. In addition, the introduction of new production processes or modification of processes often triggers subsequent disruptions associated with non-cyclical or unforeseen events [34], such as the production of personalized orders or changes in suppliers. As a result, enterprises face limited flexibility in production and operations, making it difficult to implement strategic decisions such as adjusting production models, upgrading export product structures, and changing export destination countries. Consequently, this diminishes their ability to recover exports effectively.

Secondly, digital technological diversity also leads to resource inertia of enterprises, which reduces the utilization efficiency of production resources and increases export operation costs. In a dynamic environment, enterprises must rapidly mobilize and reorganize resources to deal with adverse situations [35]. However, digital technological diversity may disperse enterprises resources and managerial attention, resulting in insufficient support for each department’s core functions and hindering the acquisition of new competitive advantages [16]. In this case, it is difficult for enterprises to effectively manage and operate various areas, and the efficiency of resource utilization is reduced, which weakens the ability of enterprises to resist risks and crises. In addition, non-IT manufacturing enterprises usually do not focus on digital technology research and development, and often lack relevant professional knowledge and management capabilities [36], such as Gen-AI applications, adaptive machine learning systems, hybrid cloud platform integration, etc. With the adoption of more kinds of digital technologies by enterprises, there are more differences in their digital knowledge. Enterprises must invest more effort and resources to understand and integrate these technologies [37], leading to higher management costs and complexity in technology recombination [38]. This, in turn, restricts the adjustments of production models and the upgrading of export product structures, weakening the enterprises’ export recovery ability. Therefore, the following hypothesis is made:

H1. 

Digital technological diversity negatively impacts enterprise export resilience.

2.2. The Mediation Effect of Supply Chain Finance

2.2.1. The Effect of Digital Technological Diversity on Supply Chain Finance

Digital technological diversity may reduce the financing ability of enterprises supply chain. Firstly, the application of multiple digital technologies increases the complexity of enterprise information processing, which in turn exacerbates supply chain information asymmetry and hinders firms’ access to supply chain finance. Existing studies believe that the application of information technology tends to create separate systems rather than reciprocal and connected systems [33]. This means that enterprises adopting multiple digital technologies simultaneously face increased complexity in analyzing heterogeneous information and managing multi-system interfaces [39]. Such complexity impedes the vertical integration of internal information and further exacerbates the information asymmetry between enterprises and financial service institutions. In addition, technology integration of all enterprises in the supply chain is essential for enterprises that want to finance through the supply chain. Nonetheless, the absence of compatibility among various digital technologies and platforms, along with the inadequate integration of systems throughout the entire chain, facilitates the formation of data islands. This undermines the speed and transparency of real-time information exchange between enterprises and financial service providers [40], heightening the uncertainty in supply chain finance. For example, Sundarakani et al. [41] found that the simultaneous application of big data analysis and blockchain would not only bring risks but also fail to create value.

Secondly, digital technological diversity disperses enterprises’ technological resources, which not only increases financing costs but also reduces the quality of supply chain information. As digital technological diversity increases, enterprises face growing constraints on both soft and hard resources, such as technical expertise and digital infrastructure, since managing diverse digital applications across internal business functions and inter-organizational boundaries demands specialized technical personnel and robust system support [33]. Therefore, diversified digital technology deployment may increase the difficulty for enterprises to fully adopt each digital technology, resulting in a negative impact on both the speed and accuracy of supply chain information delivery [42]. These shortcomings may damage the organization’s information processing ability, thus affecting its ability to meet the information visibility requirements of supply chain finance. In addition, to address the shortage of technical resources, enterprises must not only invest in employee and managerial training on digital technologies but also upgrade related hardware infrastructure [43]. These efforts increase financing costs and hinder the development of supply chain finance. Therefore, the following hypothesis is made:

H2. 

Digital technological diversity negatively impacts supply chain finance.

2.2.2. The Effect of Supply Chain Finance on Enterprise Export Resilience

Supply chain finance can not only reduce the information asymmetry in the supply chain but also alleviate financing constraints and provide assistance in stabilizing the export business of enterprises. Supply chain finance effectively integrates the logistics, information, and funds of enterprises in the supply chain, allowing financial service providers to master all detailed transaction records of financing enterprises and the business credits of other supply chain participants [44]. Through a range of schemes such as receivable/payable transfers, closed-loop systems, behavioral management, and outcome control, supply chain finance substantially reduces information asymmetry among supply chain finance participants [45,46]. Enhancing enterprise visibility enables financial service providers to identify potential risks promptly, allocate resources more effectively to mitigate supply chain disruptions, and foster stable cooperative relationships between firms and their supply chain partners, thereby improving firms’ export resilience. In addition, supply chain finance transforms the traditional credit assessment model based solely on single enterprises by allowing firms to leverage the creditworthiness of core enterprises within the supply chain. This reduces the perceived risk of financial institutions, enhances risk control, and ultimately enables firms to obtain financing at lower costs, thereby alleviating financial constraints [47]. With sufficient capital resources, enterprises are more likely to maintain or resume normal operations in the event of export business interruption. A typical example is that supply chain finance has played an important role in the resumption of enterprises’ operations in the context of COVID-19 [48].

The export costs of enterprises usually have significant prepayment characteristics, which makes them more vulnerable to the risk of a broken financial chain during operation [11]. To ensure the sustained and stable operation of the enterprise, it is necessary to maintain sufficient internal liquidity reserves. However, export enterprises are more susceptible to credit defects [49], which increase their financing difficulties. The combination of digital technology and enterprise practice not only enhances the ability of export enterprises to capture, restructure, and allocate internal resources in external market opportunities [50] but also provides them with new financing channels [9]. However, digital technological diversity not only hinders the adjustment of export models and increases export costs for enterprises but also increases information asymmetry between enterprises and financial service providers, reinforcing the financing constraints faced by enterprises, and making it difficult for shocked enterprises to recover from exporting. Therefore, the following hypothesis is made:

H3. 

Supply chain finance positively impacts enterprise export resilience.

H4. 

Supply chain finance significantly mediates the relationship between digital technological diversity and enterprise export resilience.

2.3. The Moderation Effect of Digital Interlock

Digital interlock can alleviate the inhibitory effect of digital technological diversity on supply chain finance. Firstly, digital interlock enriches the enterprises’ digital technology knowledge, and to a certain extent, it can overcome the complexity of information processing problems caused by diversified digital technology applications. Optimizing enterprise production processes and improving operational efficiency through digital technology not only depends on explicit specialized knowledge of digital technology but also on implicit digital knowledge within the enterprise [51]. Digital interlock among corporate directors helps directors break down communication barriers with external corporate members to gain more specialized digital technology knowledge [52]. In addition, directors interlocked with digital transformation firms can also obtain background knowledge about the deployment of digital technology applications, digital strategic plans, and internal information about digital technology changes, which are more important for optimizing, upgrading, and adapting digital technologies [15]. Therefore, digital interlock facilitates the accumulation of a rich knowledge base, perspectives, and perceptions of digital technology. This enables enterprises to overcome cognitive barriers in applying diversified digital technologies, enhances the compatibility of digital technology and business operations, and strengthens employees’ collective thinking about digital technology. Consequently, it improves vertical integration abilities, reduces internal information processing complexity, and promotes supply chain finance.

Secondly, digital interlock brings technological coordination and resource allocation optimization, alleviating the dispersion of technological resources caused by digital technological diversity. Specifically, by establishing digital interlock, directors can gain in-depth insights into the dynamic trends of digital technology development, including technological evolution paths, market patterns of emerging dominant platforms, and the potential value of different technological trajectories [53]. This systematic cognition enables boards to formulate more forward-looking technology investment strategies, guiding management to prioritize the allocation of core resources to digital technology with technological compatibility and interface synergy [54]. This strategic choice not only avoids investment risks caused by sudden changes in the external technological environment (such as the emergence of disruptive technologies), ensuring the effectiveness and sustainability of technology investment [55], but also strengthens the digital technology interconnection between enterprises and their supply chain partners [56], enhances overall synergy efficiency, and creates stronger conditions for the implementation of supply chain finance. Therefore, the following hypothesis is made:

H5. 

Digital interlock negatively moderates the relationship between digital technological diversity and supply chain finance.

3. Materials and Methods

3.1. Data

This study examines the impact of digital technological diversity on enterprise export resilience, the mediation effect of supply chain finance, and the moderation effect of digital interlock, using the data from Chinese listed export manufacturing companies from 2020 to 2023. At the beginning of 2020, the COVID-19 epidemic broke out in China. Restricted by the policy blockade, it was difficult for factories to resume production, leading to disruptions of manufacturing enterprises. China is regarded as the “world’s factory”, and its manufacturing industry plays a key role in the global flow of goods [57], and the disruption of China’s manufacturing industry has affected the stable operation of global supply chains. At the same time, driven by the “Internet Plus” plan, Chinese manufacturing companies have been developing digital transformation, integrating digital technologies into their production and business processes. Moreover, unlike traditional crisis events, the disruptive impact of COVID-19 covers all supply chain members and has significant implications for supply chain operations [58]. The impact of COVID-19 on China’s manufacturing industry provides a unique background for understanding how diversified digital technology applications affect enterprise export resilience in a crisis environment.

The data on digital technology, export sales, and governance structure were sourced from China Stock Market and Accounting Research database (CSMAR), and digital patent data were sourced from Chinese Research Data Services Platform (CNRDS). The foreign direct investment, gross national product, and producer price index data were obtained from the Statistical Yearbook of each city. In order to reduce the effect of extreme values, this study winsorized all continuous variables at the 1% level at both ends. After deleting the samples with missing main variables, 585 companies from 2020 to 2024 were finally obtained, with a total of 2015 enterprise-year samples.

3.2. Measures

3.2.1. Dependent Variable

Referring to Ferri et al. [11], this paper measures enterprise export resilience (EER) by the actual growth rate of enterprise export revenue compared with the export revenue in 2019, through which the index can measure the ability of enterprises to maintain the stability of export performance in market fluctuations. The larger the relative growth rate, the greater the stability of enterprise export performance under external shocks, that is, the higher the export resilience. Taking 2019 as the base year, enterprise export revenue is deflated using the production price index of each city [59], and the actual export revenue is obtained.

3.2.2. Independent Variable

Following the established methodology of Liu et al. (2023) and Kahiluoto et al. (2020) in their studies on digital technologies and diversity [16,60], digital technological diversity (DTD) is calculated by the Shannon–Weaver diversity index formula. The specific indicators are derived from the word frequency of keywords related to digital technology identified in corporate annual reports. The formula is as follows: $\mathrm{H}$ refers to digital technological diversity, $f_i$ is the word frequency of the digital technology $\mathrm{i}$, and $\mathrm{F}$ is the total word frequency of all digital technologies in an enterprise. In order to facilitate the interpretation of data on a linear scale, this study takes an exponential calculation for $\mathrm{H}$, and if the enterprise does not adopt digital technology, then the DTD is zero.

$$\mathrm{H}=-{\displaystyle \sum _{i=1}^n}\left({\displaystyle \frac{f_i}{F}}\times ln{\displaystyle \frac{f_i}{F}}\right)$$

$$\mathrm{D}\mathrm{T}\mathrm{D}=e^H$$

3.2.3. Mediator Variable

With the help of internet trading platforms, buyers, sellers, and supply chain finance service providers are connected to each other, which provides financing solutions for enterprises by optimizing inventory management, process management, and cash management. Referring to Zhu et al. [61] and Shi et al. [62], this study measures the supply chain finance (SCF) capability of a corporation by the ratio of year-end accounts payable, notes payable, and customer prepayments to total assets.

3.2.4. Moderator Variable

If there is at least one director serving in two enterprises at the same time, it indicates that the enterprises are interlocked with each other. By analyzing the list of enterprises’ board members, the interlocking network of listed manufacturing enterprises is constructed, and the digital invention patent of interlocked enterprises is identified using the “Key Digital Technology Patent Classification System (2023)” published by China Intellectual Property. Referring to Wang et al. [15], the ratio of the number of digital patents granted by interlocked enterprises to the total number of interlocked enterprises is used to measure digital interlock (DI).

3.2.5. Control Variables

Referring to Majocchi et al. [63], Dong et al. [64], and Lin et al. [65], this study controls for factors that may affect enterprise export resilience: (1) Firm size (Size): The number of employees is taken as logarithmic. The larger the firm size, the more competitive it is in the international market and able to resist risks. (2) Firm age (Age): The difference between the observation year and the establishment year is taken as a logarithm. The older the firm is, the more experience it has accumulated, and the greater the possibility of successfully responding to crises. (3) State-owned enterprises (SOE) are assigned a value of 1, while others are assigned a value of 0. State-owned enterprises are more likely to follow the government’s objectives and be influenced by policies. (4) TMT overseas experience (TMT-OE): The ratio between executives with overseas experience and the total number of executives in the enterprises. An executive team with overseas experience can reduce uncertainty in international business. (5) Research and development intensity (RDI): The ratio between R&D expenditures and operating revenues. Enterprises focused on production R&D are more inclined to generate new ideas and solutions to cope with crisis. (6) Supplier concentration (SC): The proportion of procurement amount of the top five suppliers to the procurement amount of the enterprise. An enterprise with diversified suppliers has more choices for purchasing production materials, which can resist production interruption crises. (7) Customer concentration (CC): The proportion of sales revenue of the top five customers to the total sales revenue of the enterprise. The higher the customer concentration, the more difficult it is to recover from a supply chain disruption. (8) Industry concentration (IC) reflects the degree of market competition and monopoly in the industry, and the higher the index, the stronger the systematic risk that enterprises may face in their exports. (9) Geographical position (GP): If the enterprise is located in an economic zone or an open coastal city, it is assigned a value of 1; otherwise, it is assigned a value of 0. (10) International openness (IO) is measured by the ratio of foreign direct investment to GDP in the city where the enterprise is located. The more open the city is, the more it can bring advanced overseas technology and management experience to the enterprise. After the Hausman test, this study determined to use a two-way fixed effects (year and firm effects) model for the analysis.

Table 1. Definitions and measurement of variables.

Variables	Code	Definition and Operationalization	Sources
Dependent variable	EER	EER is the actual growth rate of a company’s export revenue in year i compared to its export revenue in 2019.	CSMAR and Statistical Yearbook
Independent variable	DTD	The formula for the types of digital technologies applied by enterprises is as follows: $e^{-{\displaystyle {\sum }_{i=1}^n}\left({\displaystyle \frac{f_i}{F}}\times ln{\displaystyle \frac{f_i}{F}}\right)}$	CSMAR
Mediator variable	SCF	SCF is measured by the ratio of year-end accounts payable, notes payable, and customer prepayments to total assets.	CSMAR
Moderator variable	DI	DI is measured by the ratio of the number of digital patents granted by chain enterprises to the total number of chain enterprises.	CSMAR and CNRDS
Control variables	Size	The number of employees is taken as logarithmic.	CSMAR
	Age	The difference between the observation year and the establishment year is taken as a logarithm.	CSMAR
	SOE	Whether the enterprise is state-owned. The dummy variable is coded as 1 if it is a state-owned enterprise, otherwise 0.	CSMAR
	TMT-OE	The ratio between executives with overseas experience and the total number of executives in the enterprises.	CSMAR
	RDI	The ratio between R&D expenditures and operating revenues.	CSMAR
	SC	The proportion of procurement amount of the top five suppliers to the procurement amount of the enterprise.	CSMAR
	CC	The proportion of sales revenue of the top five customers to the total sales revenue of the enterprise.	CSMAR
	IC	IC reflects the degree of market competition and monopoly in the industry, measured by the Herfindahl index.	CSMAR
	GP	If the enterprise is located in an economic zone or an open coastal city, it is assigned a value of 1; otherwise 0.	CSMAR
	IO	IO is measured by the ratio of foreign direct investment to GDP in the city where the enterprise is located.	Statistical Yearbook

shows the definitions, measurements, and data sources of variables.

Table 2. Correlations matrix of variables.

	Size	Age	SOE	TMT-OE	RDI	SC	CC	IC	GP	IO	EER	DTD	SCF	DI
Size	1
Age	0.214 ***	1
SOE	0.218 ***	0.281 ***	1
TMT-OE	0.009	−0.067 ***	−0.108 ***	1
RDI	−0.139 ***	−0.124 ***	−0.171 ***	0.015	1
SC	−0.337 ***	−0.099 ***	−0.023	−0.030	−0.039 *	1
CC	−0.243 ***	−0.164 ***	−0.044 **	0.012	0.107 ***	0.244 ***	1
IC	0.027	−0.048**	−0.010	−0.022	−0.211 ***	0.101 ***	0.022	1
GP	−0.017	0.095 ***	−0.057 **	0.015	0.105 ***	−0.010	0.022	−0.042 *	1
IO	0.007	−0.032	−0.118 ***	0.003	0.069 ***	−0.021	−0.001	−0.001	0.286 ***	1
EER	0.025	−0.105 ***	−0.048 **	0.014	0.029	−0.016	0.013	0.000	−0.101 ***	−0.047 **	1
DTD	0.148 ***	−0.030	−0.067 ***	0.037	0.252 ***	−0.192 ***	−0.036	−0.075 ***	0.106 ***	0.135 ***	−0.127 ***	1
SCF	0.265 ***	0.025	0.097 ***	−0.008	−0.098 ***	−0.193 ***	0.151 ***	−0.057 **	−0.097 ***	−0.023	0.100 ***	0.145 ***	1
DI	0.162 ***	0.083 ***	0.050 **	0.062 ***	0.012	−0.061 ***	−0.021	−0.026	0.060 ***	−0.001	−0.022	0.064 ***	0.062 ***	1
Mean	8.004	3.053	0.265	0.005	4.881	30.956	31.158	0.089	0.289	0.038	0.253	2.431	0.15	0.155
SD	1.125	0.27	0.441	0.016	3.611	16.239	19.771	0.076	0.453	0.049	1.704	1.914	0.089	0.431
Min	5.753	2.197	0	0	0.18	7.57	4.29	0.023	0	0	−0.973	0	0.012	0
Max	10.842	3.611	1	0.077	23.47	79.67	87.82	0.444	1	0.216	21.262	8.152	0.445	3.51

Note: Firm size (Size); Firm age (Age); State-owned enterprises (SOE); TMT overseas experience (TMT-OE); R&D intensity (RDI); Supplier concentration (SC); Customer concentration (CC); Industry concentration (IC); Geographical position (GP); International openness (IO); Enterprise export resilience (EER); Digital technological diversity (DTD); Supply chain finance (SCF); Digital interlock (DI); *** p < 0.01, ** p < 0.05, * p < 0.1.

reports descriptive statistics and correlation analyses for all variables, with correlation coefficients between variables less than 0.4 and VIF values less than 2 for each variable.

4. Results

4.1. Main Effect

Table 3 reports the regression results for testing hypotheses. Model 1 tests the relationship between DTD and EER without adding control variables, while Model 2 tests this relationship with adding control variables. The regression results show that with the addition of control variables, the adjusted R-square of Model 2 is much larger than that of Model 1, and that DTD ($\mathsf{\beta }=-0.0589,p<0.05$) still significantly inhibits EER, indicating the validity of H1.

4.2. Mediation Effect

Model 3 shows that DTD ($\mathsf{\beta }=-0.0024,p<0.001$) significantly inhibits SCF, and Model 4 shows that SCF ($\mathsf{\beta }=1.8071,p<0.001$) significantly enhances EER. At the same time, compared to the estimated coefficient of DTD in Model 2, the estimated coefficient of DTD in Model 4 ($\mathsf{\beta }=-0.0546,p<0.05$) decreased, indicating that SCF plays a partial mediating role. In addition, it can be concluded that the direct effect of DTD on EER is −0.0546, and the indirect effect is −0.0043 (−0.0024 $\times$ 1.8071 = −0.0043), indicating that DTD can indirectly inhibit EER through SCF. H2, H3, and H4 are validated.

4.3. Moderation Effect

In order to reduce the impact of collinearity, this study centralized DTD and DI before calculating the interaction term DTD $\times$ DI. Model 5 in Table 2 shows that the coefficient of the interaction term DTD $\times$ DI is significantly positive at the 5% significance level, indicating that DI significantly alleviates the inhibitory effect of DTD on SCF, and H5 is verified.

Table 3. Regression results of main effect, mediation effect, and moderation effect.

Variables	Enterprise Export Resilience		Supply Chain Finance	Enterprise Export Resilience	Supply Chain Finance
	Model 1	Model 2	Model 3	Model 4	Model 5
Control variables
Firm size		0.8999 ***	0.0255 ***	0.8538 ***	0.0256 ***
		[0.3225]	[0.0085]	[0.3200]	[0.0085]
Age		−2.6339 *	−0.0487	−2.5459 *	−0.0468
		[1.5093]	[0.0556]	[1.5092]	[0.0557]
State-owned enterprises		−0.3360 *	0.0051	−0.3452 *	0.0051
		[0.1744]	[0.0076]	[0.1784]	[0.0076]
TMT overseas experience		2.0248	−0.0035	2.0311	−0.0056
		[1.4788]	[0.0538]	[1.4753]	[0.0539]
R&D intensity		−0.0737 *	−0.0009	−0.0721 *	−0.0009
		[0.0425]	[0.0009]	[0.0423]	[0.0009]
Supplier concentration		−0.0012	−0.0002	−0.0009	−0.0002
		[0.0042]	[0.0002]	[0.0042]	[0.0002]
Customer concentration		−0.0103	0.0004 **	−0.0110	0.0004 **
		[0.0111]	[0.0002]	[0.0111]	[0.0002]
Industry concentration		−0.1231	−0.0552 **	−0.0233	−0.0545 **
		[0.4962]	[0.0214]	[0.4946]	[0.0215]
Geographical position		−2.9398 ***	0.0264 ***	−2.9875 ***	0.0306 ***
		[0.1648]	[0.0079]	[0.1674]	[0.0085]
International openness		1.3264 *	0.0215	1.2876 *	0.0192
		[0.7013]	[0.0321]	[0.7034]	[0.0320]
Independent variables
Digital technological diversity (DTD)	−0.1134 ***	−0.0589 **	−0.0024 ***	−0.0546 **	−0.0022 ***
	[0.0184]	[0.0229]	[0.0008]	[0.0231]	[0.0008]
Mediator variable
Supply chain finance				1.8071 ***
				[0.6443]
Moderator variable
Digital interlock (DI)					−0.0024
					[0.0047]
Interactions
DTD $\times$ DI					0.0059 **
					[0.0026]
Constant	0.5284 ***	2.8426	0.0929	2.6747	0.0842
	[0.0765]	[3.8671]	[0.1699]	[3.8853]	[0.1703]
Model fit
Observations	2015	2015	2015	2015	2015
F-value	38.1317	441.1575	19.2144	407.9589	16.4329
R-squared	0.0157	0.7903	0.8880	0.7912	0.8880
Firm FE	Yes	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes	Yes

Note: Standard errors in brackets; * p < 0.1, ** p < 0.05, *** p < 0.01.

Table 3. Regression results of main effect, mediation effect, and moderation effect.

Variables	Enterprise Export Resilience		Supply Chain Finance	Enterprise Export Resilience	Supply Chain Finance
	Model 1	Model 2	Model 3	Model 4	Model 5
Control variables
Firm size		0.8999 ***	0.0255 ***	0.8538 ***	0.0256 ***
		[0.3225]	[0.0085]	[0.3200]	[0.0085]
Age		−2.6339 *	−0.0487	−2.5459 *	−0.0468
		[1.5093]	[0.0556]	[1.5092]	[0.0557]
State-owned enterprises		−0.3360 *	0.0051	−0.3452 *	0.0051
		[0.1744]	[0.0076]	[0.1784]	[0.0076]
TMT overseas experience		2.0248	−0.0035	2.0311	−0.0056
		[1.4788]	[0.0538]	[1.4753]	[0.0539]
R&D intensity		−0.0737 *	−0.0009	−0.0721 *	−0.0009
		[0.0425]	[0.0009]	[0.0423]	[0.0009]
Supplier concentration		−0.0012	−0.0002	−0.0009	−0.0002
		[0.0042]	[0.0002]	[0.0042]	[0.0002]
Customer concentration		−0.0103	0.0004 **	−0.0110	0.0004 **
		[0.0111]	[0.0002]	[0.0111]	[0.0002]
Industry concentration		−0.1231	−0.0552 **	−0.0233	−0.0545 **
		[0.4962]	[0.0214]	[0.4946]	[0.0215]
Geographical position		−2.9398 ***	0.0264 ***	−2.9875 ***	0.0306 ***
		[0.1648]	[0.0079]	[0.1674]	[0.0085]
International openness		1.3264 *	0.0215	1.2876 *	0.0192
		[0.7013]	[0.0321]	[0.7034]	[0.0320]
Independent variables
Digital technological diversity (DTD)	−0.1134 ***	−0.0589 **	−0.0024 ***	−0.0546 **	−0.0022 ***
	[0.0184]	[0.0229]	[0.0008]	[0.0231]	[0.0008]
Mediator variable
Supply chain finance				1.8071 ***
				[0.6443]
Moderator variable
Digital interlock (DI)					−0.0024
					[0.0047]
Interactions
DTD $\times$ DI					0.0059 **
					[0.0026]
Constant	0.5284 ***	2.8426	0.0929	2.6747	0.0842
	[0.0765]	[3.8671]	[0.1699]	[3.8853]	[0.1703]
Model fit
Observations	2015	2015	2015	2015	2015
F-value	38.1317	441.1575	19.2144	407.9589	16.4329
R-squared	0.0157	0.7903	0.8880	0.7912	0.8880
Firm FE	Yes	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes	Yes

Note: Standard errors in brackets; * p < 0.1, ** p < 0.05, *** p < 0.01.

4.4. Endogeneity and Robustness Checks

The application of digital technology by enterprises may be influenced by factors such as enterprise characteristics, production models, and policies, and is not random. Therefore, this study adopts the propensity score matching method to alleviate the problem of self-selection bias. Firstly, based on the industry-year mean of DTD, those above the average are considered the treatment group, while those below the average are considered the control group. Secondly, combining all control variables, industry and year dummy variables, radius (caliper = 0.05) matching is performed. The matching results satisfy the balance test requirements, and there is no significant difference between the matched variables. Regression analysis is conducted again using the matched samples, and the regression results are shown in

Table 4. Post-match diagnostic regressions.

Variables	Enterprise Export Resilience	Supply Chain Finance	Enterprise Export Resilience	Supply Chain Finance
	Model 1	Model 2	Model 3	Model 4
Control variables
Firm size	0.8911 ***	0.0252 ***	0.8492 ***	0.0253 ***
	[0.3244]	[0.0085]	[0.3224]	[0.0085]
Age	−2.7622 *	−0.0531	−2.6736 *	−0.0518
	[1.5241]	[0.0565]	[1.5250]	[0.0566]
State-owned enterprises	−0.3400 *	0.0022	−0.3436 *	0.0022
	[0.1766]	[0.0063]	[0.1806]	[0.0063]
TMT overseas experience	1.9025	−0.0021	1.9060	−0.0040
	[1.4951]	[0.0545]	[1.4916]	[0.0546]
R&D intensity	−0.0740	−0.0009	−0.0725	−0.0009
	[0.0464]	[0.0010]	[0.0462]	[0.0010]
Supplier concentration	−0.0010	−0.0002	−0.0006	−0.0002
	[0.0045]	[0.0002]	[0.0045]	[0.0002]
Customer concentration	−0.0107	0.0004 **	−0.0113	0.0004 **
	[0.0116]	[0.0002]	[0.0116]	[0.0002]
Industry concentration	−0.0809	−0.0552 **	0.0112	−0.0547 **
	[0.5119]	[0.0220]	[0.5109]	[0.0220]
Geographical position	−2.9192 ***	0.0258 ***	−2.9622 ***	0.0296 ***
	[0.1699]	[0.0080]	[0.1722]	[0.0086]
International openness	1.3417 *	0.0202	1.3080 *	0.0182
	[0.7039]	[0.0322]	[0.7057]	[0.0321]
Independent variables
Digital technological diversity (DTD)	−0.0621 ***	−0.0024 ***	−0.0581 **	−0.0022 ***
	[0.0233]	[0.0008]	[0.0234]	[0.0008]
Mediator variable
Supply chain finance			1.6667 **
			[0.6557]
Moderator variable
Digital interlock (DI)				−0.0028
				[0.0047]
Interactions
DTD $\times$ DI				0.0048 *
				[0.0027]
Constant	3.2855	0.1124	3.0982	0.1057
	[3.9153]	[0.1726]	[3.9349]	[0.1730]
Model fit
Observations	1950	1950	1950	1950
F-value	421.3138	18.9936	388.8392	16.1443
R-squared	0.7931	0.8875	0.7938	0.8875
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes

Note: Standard errors in brackets; * p < 0.1, ** p < 0.05, *** p < 0.01.

. The regression coefficients and significance of the core variables have not changed significantly, still supporting the research hypotheses.

In addition, referring to Wang et al. [66] and Park & Gupta [67], this study uses Gaussian copula approach to model the joint distribution of potential endogenous variables and error terms to alleviate endogeneity. Firstly, the null hypothesis that the independent variables are normally distributed is rejected by testing the DTD (W = 0.96834, p = 0.000), which provides the possibility for further analysis. Secondly, we incorporate copula terms DTD_copula into the regression model and re-examine the research hypotheses. The regression results are shown in

Table 5. Gaussian regression results.

Variables	Enterprise Export Resilience	Supply Chain Finance	Enterprise Export Resilience	Supply Chain Finance
	Model 1	Model 2	Model 3	Model 4
Control variables
Firm size	0.8997 ***	0.0255 ***	0.8534 ***	0.0256 ***
	[0.3226]	[0.0085]	[0.3201]	[0.0085]
Age	−2.6450 *	−0.0476	−2.5585 *	−0.0455
	[1.5085]	[0.0556]	[1.5084]	[0.0557]
State-owned enterprises	−0.3342 *	0.0049	−0.3431 *	0.0049
	[0.1757]	[0.0076]	[0.1799]	[0.0076]
TMT overseas experience	2.0415	−0.0051	2.0508	−0.0056
	[1.4809]	[0.0540]	[1.4770]	[0.0542]
R&D intensity	−0.0736 *	−0.0009	−0.0720 *	−0.0009
	[0.0426]	[0.0009]	[0.0423]	[0.0009]
Supplier concentration	−0.0012	−0.0002	−0.0008	−0.0002
	[0.0042]	[0.0002]	[0.0042]	[0.0002]
Customer concentration	−0.0103	0.0004 **	−0.0110	0.0004 **
	[0.0111]	[0.0002]	[0.0111]	[0.0002]
Industry concentration	−0.1210	−0.0554 ***	−0.0204	−0.0549 **
	[0.4952]	[0.0214]	[0.4935]	[0.0214]
Geographical position	−2.9372 ***	0.0262 ***	−2.9847 ***	0.0303 ***
	[0.1654]	[0.0079]	[0.1679]	[0.0087]
International openness	1.3285 *	0.0213	1.2898 *	0.0189
	[0.7016]	[0.0321]	[0.7035]	[0.0320]
DTD_copula	−0.0049	0.0005	−0.0057	0.0005
	[0.0095]	[0.0005]	[0.0095]	[0.0005]
Independent variables
Digital technological diversity (DTD)	−0.0579 **	−0.0025 ***	−0.0534 **	−0.0023 ***
	[0.0227]	[0.0008]	[0.0228]	[0.0008]
Mediator variable
Supply chain finance			1.8149 ***
			[0.6463]
Moderator variable
Digital interlock (DI)				−0.0010
				[0.0023]
Interactions
DTD $\times$ DI				0.0059 **
				[0.0026]
Constant	2.8781	0.0895	2.7156	0.0805
	[3.8671]	[0.1700]	[3.8849]	[0.1705]
Model fit
Observations	2015	2015	2015	2015
F-value	405.0812	17.6722	377.5481	15.3286
R-squared	0.7902	0.8880	0.7910	0.8880
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes

Note: Standard errors in brackets; * p < 0.1, ** p < 0.05, *** p < 0.01.

and demonstrate statistically insignificant coefficients for DTD_copula, suggesting that potential endogeneity concerns regarding DTD are unlikely to substantially affect the research findings.

To further validate the robustness of the research results, on the one hand, referring to Liu et al. [16], the digital technology is divided into five dimensions (artificial intelligence, blockchain, cloud computing, big data, and digital technology application), recalculating the DTD, and the regression results are shown in

Table 6. Regression results for the alternative measure of digital technological diversity.

Variables	Enterprise Export Resilience	Supply Chain Finance	Enterprise Export Resilience	Supply Chain Finance
	Model 1	Model 2	Model 3	Model 4
Control variables
Firm size	0.9067 ***	0.0259 ***	0.8586 ***	0.0259 ***
	[0.3223]	[0.0085]	[0.3198]	[0.0085]
Age	−2.5275 *	−0.0456	−2.4429	−0.0449
	[1.5059]	[0.0555]	[1.5069]	[0.0555]
State-owned enterprises	−0.3307 **	0.0052	−0.3404 **	0.0051
	[0.1664]	[0.0077]	[0.1710]	[0.0076]
TMT overseas experience	1.8720	−0.0087	1.8883	−0.0052
	[1.4713]	[0.0534]	[1.4665]	[0.0536]
R&D intensity	−0.0730 *	−0.0009	−0.0714 *	−0.0009
	[0.0426]	[0.0009]	[0.0424]	[0.0009]
Supplier concentration	−0.0010	−0.0002	−0.0007	−0.0002
	[0.0042]	[0.0002]	[0.0042]	[0.0002]
Customer concentration	−0.0103	0.0004 **	−0.0110	0.0004 **
	[0.0111]	[0.0002]	[0.0111]	[0.0002]
Industry concentration	−0.1584	−0.0564 ***	−0.0535	−0.0553 **
	[0.5030]	[0.0215]	[0.4999]	[0.0215]
Geographical position	−2.9564 ***	0.0257 ***	−3.0042 ***	0.0279 ***
	[0.1634]	[0.0079]	[0.1657]	[0.0087]
International openness	1.2696 *	0.0193	1.2337 *	0.0187
	[0.6998]	[0.0323]	[0.7016]	[0.0323]
Independent variables
Digital technological diversity (DTD)	−0.3667 **	−0.0113 **	−0.3457 **	−0.0091 *
	[0.1465]	[0.0054]	[0.1472]	[0.0054]
Mediator variable
Supply chain finance			1.8583 ***
			[0.6405]
Moderator variable
Digital interlock (DI)				−0.0024
				[0.0025]
Interactions
DTD $\times$ DI				0.0498 **
				[0.0230]
Constant	2.5133	0.0806	2.3635	0.0767
	[3.8476]	[0.1698]	[3.8689]	[0.1700]
Model fit
Observations	2015	2015	2015	2015
F-value	418.0645	18.3864	388.5582	15.8032
R-squared	0.7901	0.8876	0.7910	0.8876
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes

Note: Standard errors in brackets; * p < 0.1, ** p < 0.05, *** p < 0.01.

. On the other hand, there are more digital technology applications in the digital product manufacturing industry [68]. In order to avoid the interference of these enterprises, this study excludes the samples of the enterprises in the computer, communication, and other electronic equipment manufacturing industries, and the regression results are shown in

Table 7. Regression results excluding computer, communication, and other electronic equipment enterprises.

Variables	Enterprise Export Resilience	Supply Chain Finance	Enterprise Export Resilience	Supply Chain Finance
	Model 1	Model 2	Model 3	Model 4
Control variables
Firm size	0.8784 **	0.0218 **	0.8424 **	0.0220 **
	[0.3673]	[0.0093]	[0.3645]	[0.0093]
Age	−4.6646 **	0.0185	−4.6951 **	0.0209
	[1.9731]	[0.0571]	[1.9802]	[0.0572]
State-owned enterprises	−0.3385 *	0.0050	−0.3468 *	0.0050
	[0.1806]	[0.0074]	[0.1843]	[0.0074]
TMT overseas experience	1.7921	−0.0110	1.8102	−0.0119
	[1.6387]	[0.0554]	[1.6365]	[0.0556]
R&D intensity	−0.0864 *	−0.0010	−0.0848 *	−0.0009
	[0.0478]	[0.0011]	[0.0476]	[0.0011]
Supplier concentration	−0.0024	−0.0001	−0.0023	−0.0001
	[0.0047]	[0.0002]	[0.0047]	[0.0002]
Customer concentration	−0.0104	0.0003 **	−0.0110	0.0003 **
	[0.0132]	[0.0002]	[0.0132]	[0.0002]
Industry concentration	0.0081	−0.0526 **	0.0948	−0.0516 **
	[0.5264]	[0.0214]	[0.5237]	[0.0215]
Geographical position	−2.7193 ***	0.0275 ***	−2.7646 ***	0.0314 ***
	[0.1758]	[0.0088]	[0.1784]	[0.0094]
International openness	2.1008 ***	0.0305	2.0505 ***	0.0284
	[0.6873]	[0.0345]	[0.6883]	[0.0344]
Independent variables
Digital technological diversity (DTD)	−0.0659 **	−0.0023 **	−0.0621 **	−0.0021 **
	[0.0295]	[0.0010]	[0.0297]	[0.0010]
Mediator variable
Supply chain finance			1.6480 **
			[0.7038]
Moderator variable
Digital interlock (DI)				−0.0007
				[0.0054]
Interactions
DTD $\times$ DI				0.0064 **
				[0.0031]
Constant	9.1370 *	−0.0845	9.2763 **	−0.0952
	[4.6849]	[0.1795]	[4.7223]	[0.1799]
Model fit
Observations	1762	1762	1762	1762
F-value	398.8246	18.2555	367.1793	15.5666
R-squared	0.7994	0.8906	0.8000	0.8906
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes

Note: Standard errors in brackets; * p < 0.1, ** p < 0.05, *** p < 0.01.

. It can be seen that the coefficients and significance of DTD, SCF, and DTD $\times$ DI have not been substantially changed, and the research hypotheses still hold.

5. Discussion

5.1. Theoretical Implications

Firstly, this study found that in adversity, the diversified application of digital technology is actually not conducive to establishing enterprise export resilience, which reveals the debate between digital technology and organizational resilience [3,12]. Existing research is mainly based on dynamic capability theory, resource orchestration theory, and information processing theory, analyzing the advantages and disadvantages of digital technology from the perspective of static technological capabilities [19]. This not only simplifies the dynamic capabilities of digital technology in resource acquisition and allocation but also ignores the unique role of digital technology in specific contexts. Responding to calls from Favoretto et al. [12], Liang et al. [28], and Li et al. [69], this study emphasizes how environmental adversity subverts supposed advantages and challenges global technological optimism. By taking into account the reality of nested multiple technologies, this study provides a new insight into both the strengths and weaknesses of digital technology applications.

Secondly, this study reveals the mediation role of supply chain finance in the impact of digital technology diversity on enterprise export resilience. This implies that during a crisis, digital technology diversity limits the financing and information sharing capabilities of enterprises, which is not conducive to building enterprise export resilience. Existing research on the negative impact of digital technology mainly focuses on investment costs and management expenses, emphasizing the limited internal resource reserves of enterprises [70]. This study found that in adversity, digital technology diversity limits the ability of supply chain finance, providing a novel perspective for international controversies about the dark side of digital technology, particularly important for small and medium-sized enterprises in emerging markets. Therefore, in order to understand the impact of digital technology diversity on enterprises export resilience, it is also necessary to focus on enterprises’ supply chain finance capabilities. This finding responds to Li et al. [69] call for studying how export enterprises can resist crises and how to cope with the negative risks brought by digital technology.

Finally, this study extends the literature on digital technology and supply chain finance by exploring the role of digital interlock in addressing the negative effects of digital technological diversity and elucidates the importance of enterprise digital interlock strategies, especially for non-high-tech enterprises. Specifically, digital interlock alleviates the inhibitory effects of digital technology diversity on supply chain finance. This result aligns with the existing research that increased inter-firm digital interlock facilitates overcoming the technical challenges of digital technology adoption. Such interlocks facilitate better optimization and integration of digital technology [15], thereby reducing information processing complexity.

5.2. Managerial Implications

Firstly, digital technological diversity in crisis environments is detrimental to the establishment of export resilience, which warns against the widespread one-size-fits-all digital expansion in international business. Therefore, when formulating digitalization strategies, managers need to take full account of changes in the external environment, improve the adaptability of digital technologies to the external environment, and reduce the routine inertia and resource inertia generated by diversified technology nesting.

Secondly, digital technological diversity can also hinder the establishment of enterprise export resilience by inhibiting supply chain finance capabilities. Enterprises should work together with supply chain partners to develop standardized data sharing agreements and improve information transparency among supply chain enterprises, so that supply chain finance service providers can quickly assess enterprise credit risk and help enterprises facing export disruptions obtain more financing and information channels.

Finally, inter-firm digital interlock helps firms learn external heterogeneous technological knowledge and tacit knowledge, alleviating the negative impact of digital technological diversity on supply chain finance. Therefore, members of the corporate board should focus on diversified interlocks to learn and assimilate knowledge of digital technology applications from external firms to overcome internal cognitive limitations and organizational inertia.

5.3. Limitations

Firstly, the measurement of digital technology diversity is based on a holistic perspective, but digital technology is also decomposed into digital technology preparation, exploration, and development according to different application processes [71]. Moreover, although this study considers the reality of nested digital technology, it does not delve into the dynamic changes in the application process of digital technology. Therefore, future research could consider the dynamic changes in the impact of digital technology diversity on the establishment of enterprise export resilience in different processes.

Secondly, existing research generally agrees that supply chain finance not only has the function of financing but also has the role of information transfer. Limited to the single measurement method, this study mainly measures the financing ability of enterprise supply chain finance. Future research can combine with the connotation of supply chain finance to comprehensively measure the enterprise supply chain finance ability.

Finally, the research sample is limited to Chinese export manufacturing enterprises. In the face of COVID-19, the Chinese government implemented a strict control policy, which largely restricted enterprises’ production and operation activities and likely stimulated Chinese enterprises to be more eager to apply diversified digital technologies to resist this risk. Therefore, future research can explore the relationship between digital technology diversity and enterprises export resilience using data on enterprises from different countries and incorporate country-level policy factors as boundary conditions.