A Study of the Impact of Manufacturing Input Digitization on Firms’ Organizational Resilience: Evidence from China

Abstract

In the era of digital economy, boosting investment in digitalization serves as a crucial approach for the manufacturing sector to bolster its robustness and forge competitive edges. Utilizing data from listed manufacturing firms spanning 2007–2022, this paper investigates how the digital transformation of manufacturing inputs influences the organizational resilience of businesses and the underlying mechanism. The findings indicate that investment in digital technologies significantly enhances organizational resilience. The mechanism analysis suggests that investing in digitization will enhance organizational resilience by increasing the firm’s technological innovation capacity, and thus financial redundancy positively moderates the positive impact of digitization on organizational resilience. Heterogeneity analysis shows that there is heterogeneity among different enterprise sizes and industry intensities; enterprises should adjust their digitalization investment strategies in a timely manner according to their industries and enterprise sizes to strengthen and consolidate their organizational resilience. The results of this research provide practical proof and guidance for the development of the manufacturing sector, small and medium-sized enterprises (SMEs), and the strengthening of organizational resilience.

1. Introduction

Currently, a new round of technological innovation and industrial upgrading is making swift progress, and a series of emergencies are occurring frequently, which puts organizations in a complex environment full of uncertainty, suffering impacts and threatening their own survival. The significant differences in behavioral patterns and performance levels exhibited by different enterprises undoubtedly provide us with valuable insights, namely, in the face of crisis, an enterprise’s capacity to respond and its aptitude for making strategic adjustments exert a crucial influence on its long-term development. One of the key reasons for this is the difference in organizational resilience. Due to the frequent occurrence of “black swan” events and potential “grey rhino” risks, the concept of organizational resilience, which has specific contextual significance, has re-entered the vision of management scholars and gradually become a hot topic for their in-depth research. Organizational resilience is not only a company’s capacity to handle unforeseen incidents, but also reflects their potential for sustainable development in a complex and changing environment. In such an environment full of challenges and opportunities, enterprises must have strong organizational resilience to cope with various crises and changes in order to achieve sustainable growth and prosperity.

The Outline of China’s Fourteenth Five-Year Plan explicitly indicates that the strategy of strengthening the nation through manufacturing must be thoroughly implemented, to achieve the goal of intelligentization and digitalization of the manufacturing industry. For 12 consecutive years, China’s manufacturing value added has held the top spot globally, and it boasts the most comprehensive industrial system in the world., and is the country with the most complete industrial system in the world. Accelerate the growth of the digital economy, foster the deep integration of the digital economy with the real economy, drive the digital transformation of manufacturing and other sectors, and bolster the robustness and adaptability of the industrial chain. These policies serve as both strategic guidance for the development of manufacturing and a powerful impetus for enhancing the organizational resilience of enterprises. Therefore, improving the organizational resilience of manufacturing enterprises has an important impact on responding to alterations in the intricate external environment and realizing the vision of a “powerful manufacturing nation”. In view of this, it is essential to conduct in depth the effects and mechanisms of manufacturing digitalization on organizational resilience, so as to provide empirical evidence and policy guidance for the synergistic promotion of manufacturing digitalization and the improvement of corporate organizational resilience.

Resilience originates from “resilire” and “resilio”, signifying “to rebound” or “to spring back” in Latin [1]. As time has passed, numerous disciplines have delved into resilience, encompassing psychology, organizational science, engineering, and ecology. Meyer [2] brought resilience into the research domain of organizational management, thereby opening the gateway to investigations on organizational resilience. Although scholars have yet to reach a consensus regarding organizational resilience, the majority hold that it is a robust capability associated with successfully absorbing, adapting to, and ultimately utilizing disruptive contingencies that might pose a threat to survival [1]. In recent years, with the burgeoning interest in organizational resilience, the literature on it has been on the rise [3,4,5,6]. Research on the impact of firms’ organizational resilience of firms has focused on two main areas: internal and external. Internally, management style affects the effectiveness of corporate governance and the level of decision-making, which in turn affects organizational resilience [7], and firms can indirectly enhance their resilience to risks and provide strong support for resilience through the rational use of human capital and the promotion of knowledge and technology accumulation [8]. Firms with higher levels of digitization before the crisis then showed higher resilience in a pandemic [9]. When considering external factors, maintaining a positive attitude during a crisis can help firms address the impacts of the crisis through creative ideas, ultimately enhancing their performance and survival rate [10]. The experience acquired from crises heightens the probability of taking positive predictive measures and simultaneously diminishes the likelihood of adopting an entirely passive strategy to respond to adversity [11]. Organizational resilience is also promoted under the influence of external contexts of public health events, such as new crown epidemics or extreme weather [12,13,14,15].

In the realm of research on the digitalization of the manufacturing sector, scholars formerly contended that digital transformation entails enhancing production efficiency and enterprise performance via digital technologies [16], or leveraging information technologies such as ERP [17] to elevate management and decision-making standards. Presently, attention is primarily focused on the enterprise and industry levels. On the enterprise side, digitalization positively affects the upgrading of enterprise value chains [18], and digitalization serves as a crucial means for enterprises to enhance their involvement in international trade and boost their trade earnings [19]. Digital technology applications play an important role in achieving net-zero emissions in the manufacturing industry [20], and can facilitate the green transformation of manufacturing enterprises [21]. At the industry level, studies have found that upgrading the global manufacturing value chain by investing in digitalization and breaking through traditional trade can promote higher labor levels, product upgrading, and the creation of new factors of production [22,23], and the development of digitalization can optimize services and thereby reduce the transaction costs of the value chain [24], increase enterprise skill premiums and wage levels [25], and enhance the connectivity of domestic production networks [26].

In conclusion, this article conducts empirical analysis on the influence of digital investment on organizational resilience by selecting the data of manufacturing listed companies from 2007 to 2022. The marginal contributions of this paper may lie in the following three aspects. Firstly, referring to the latest achievements of digital investment in the manufacturing sector, the input-output method is employed to calculate the digital investment index. Growth and volatility are utilized to represent the organizational resilience of the manufacturing industry, facilitating the analysis and accurate revelation of the impact of digital investment on organizational resilience. Secondly, it analyzes and verifies the mechanism through which digitalization enhances enterprise organizational resilience by promoting technological innovation and the moderating effect of financial redundancy, which is conducive to a deeper understanding of the positive impact and development path of the digitalization level. Third, the heterogeneous impact of digitalization on organizational toughness is explored with respect to the nature of enterprise ownership, size, and industry characteristics, revealing the deep connection between digitalization and the development of organizational toughness of real enterprises, which provides solid theoretical support and decision-making references for the formulation and precise implementation of relevant policies.

2. Theoretical Analysis and Research Hypotheses

2.1. Impact of Manufacturing Digitization on Organizational Resilience

Under the background of the booming development of the digital economy, as information technology advances rapidly and production factors continue to be innovated, digital technology has been comprehensively integrated into the optimization of organizational structure, the reshaping of processes, and the innovation of activity configurations, and has emerged as a pivotal force driving the development of organizational resilience. Digitalization helps enterprises to conduct data mining, analysis, and processing, so as to rapidly identify and perceive external risk signals and improve reaction sensitivity [27], in order to help enterprises to effectively identify crisis events and quickly respond to unexpected crises. Enterprises should actively explore and fully utilize the capabilities of digital technology to continuously enhance their organizational resilience by integrating resources to cope with various challenges due to crises [28]. Enterprise digitization empowers organizations with new vitality and momentum when dealing with uncertain environments and crisis situations, and improving enterprise digitization reduces the complexity and uncertainty of the organization’s information, thus empowering the organization to recover and bounce back to adapt to risky environments under the impact of unfavorable events [12]. Digital capabilities will increase the ability of firms to adapt to adversity, helping them to remove crises and enhance their operations to help them recover better when their existing resources are unable to cope with new problems [29]. Through the effective use of digital means, enterprises can significantly improve the comprehensiveness of their information construction, enhance the effectiveness of control and monitoring of production processes, and strengthen the risk defense system [30]. The development of digital technology promotes the use of more digital information technology, by reducing internal and external costs [31], improving the efficiency of resource utilization, helping enterprises to create a unique core competitiveness resources, to ensure that they maintain an advantageous position in the fierce market competition, and to enhance the resilience and toughness of the enterprise. In the process of realizing the enhancement of organizational resilience, digitalization plays a key role, and the combination of big data and digital technology is a key link in realizing organizational resilience.

Accordingly, this paper proposes hypothesis H1: Digitalization of manufacturing industry can enhance the organizational resilience of enterprises.

2.2. Manufacturing Digitalization, Technological Innovation and Organizational Resilience

The utilization of digital technology can reinforce the informatization process and augment the technological innovation capability of enterprises [32,33]; at the same time, some studies have shown that innovation provides enterprises with the ability to maintain stability and flexibility in the face of external contingencies [34], and through the introduction of new technologies to develop new products and innovative business models, which not only shortens the response time to deal with emergencies [35], but also provides effective solutions for enterprises to deal with crises [1] and ensures the sustainable and stable development of enterprises. In addition, digitalization promotes the collection of information from various links within the enterprise and realizes the comprehensive refinement of product production and management, thus reducing costs in the procurement, manufacturing, and sales processes [36] and improving production efficiency [37]. Manufacturing enterprises with technological innovation as the core driving force can not only enhance their own development effectiveness, but also significantly improve market competitiveness and strengthen their organizational resilience. The enhancement of enterprise technological capabilities can promote the enhancement of organizational resilience [38], and enterprises with strong technological innovation capabilities are able to maintain a higher level of resilience in the aftermath of shocks [39]. Digital technological innovation is an important driver of resilience in adversity [40]. At the same time, innovation can also strengthen the core competitiveness of enterprises, results in coping with external environmental pressures [41], and help enterprises avoid risks and achieve sustainability [42]. Enterprises ought to fully leverage the positive influence of digitalization on organizational resilience, probe into the risk prediction potential of digital technology, and facilitate innovation activities, thereby fully exploiting the effect of digitalization in cultivating resilience.

In summary, this paper proposes hypothesis H2: Digitalization of the manufacturing industry can enhance the technological innovation level of enterprises, which in turn strengthens the organizational resilience of enterprises.

2.3. Manufacturing Digitalization, Financial Redundancy and Organizational Resilience

Financial redundancy is defined as liquidity and risk-free borrowing capacity in excess of a firm’s existing operating and debt needs, and is excess financial resources over and above existing business needs. As an important component of organizational redundancy, financial redundancy is a non-sinkable resource that is highly flexible and easy for firms to quickly utilize and allocate [43]. As a hard capability in the formation of organizational resilience, financial redundancy is an important moderating variable. Specifically, financial redundancy provides enterprises with the necessary resource support for digital transformation and can mitigate or flush the risk of uncertainty brought about by the mid-transformation, thus reinforcing the positive impact of digital transformation on organizational resilience. First, financial redundancy can support enterprises to try new strategies and carry out breakthrough innovations [44], which is conducive to enterprises’ ability to improve innovation and enhance organizational resilience on the basis of digital transformation [45]. At the same time, as a financial resource beyond business needs, financial redundancy can be flexibly deployed in the integration and upgrading of digital technologies or platforms, thus enhancing the effectiveness of digital transformation in monitoring and forecasting, coordination and integration, and learning. Secondly, financial redundancy has a certain buffer role in times of crisis, enterprises have the ability to quickly adapt and change, which helps to cope with environmental changes and crisis events [46], the existence of financial redundancy makes organizational resilience to a certain extent, in the face of the risks posed by the digital transformation, the enterprise is able to maintain a better ability to develop in the face of the impact [47].

Therefore, this paper proposes hypothesis H3: Financial redundancy reinforces the positive impact of digitalization on organizational resilience.

3. Methodology, Variable Description, and Data

3.1. Benchmark Regression Model

To test the relevant hypotheses, this paper constructs a benchmark regression model set as follows:

$$Re{s}_{i,t}={\beta }_0+{\beta }_{1}Di{g}_{i,t}+{\beta }_{2}Control{s}_{i,t}+{\lambda }_{i,t}+{\epsilon }_{i,t}$$

(1)

In the model, $\mathrm{i},\mathrm{t}$ represent firms and years, respectively. The explanatory variable $Re{s}_{i,t}$ denotes the strength of organizational resilience of the firm, the explanatory variable $Di{g}_{i,t}$ denotes the level of digitization of the firm, $Control{s}_{i,t}$ denotes a series of control variables, ${\lambda }_{i,t}$ denotes individual and year double fixed effects, and ${\epsilon }_{i,t}$ is a random error term.

3.2. Variable Descriptions

3.2.1. Dependent Variables

The explanatory variable is organizational resilience. Measurement approaches in the existing literature are mainly categorized into direct and indirect measures. Direct measures such as Liang [48] classified organizational resilience into the ability to withstand risk, adapt to adjust and recover to overtake; Patriarca et al. [49] unfolded it from the four dimensions of monitoring, reacting, predicting, and learning analysis, and Lu et al. [50] evaluated the three dimensions of stability, sensitivity, and synergy. Indirect measures are mainly measured by the financial data of the firm, such as Ortiz [51] by financial volatility, sales growth rate, and survival rate; and DesJardine et al. [52] categorize it based on the duration of the stock price decline period, the magnitude of the decline, and the degree of recovery of the stock price. Baghersad [53] assesses it through initial loss, maximum loss, and cumulative loss over time. In this paper, from the perspective that organizational resilience enables firms to remain well and sustainable amidst shocks, we adopt Ortiz and Bansal’s approach and measure it in terms of two indicators: growth and volatility. Growth is measured by the cumulative increase in operating income in three years, expressed as growth; volatility is measured by the standard deviation of stock returns in each month of the year, expressed as sd. Finally, the entropy weighting method is utilized to calculate the “organizational resilience” variable, which is operated as follows:

(1)

Data normalization. The polar value method is used to normalize the data, adjusting the value of each indicator to the range of 0 to 1, so as to reduce the error due to the different numerical units. i denotes the enterprise, j denotes the growth indicator and the volatility indicator, and $x_{ij}^{\prime }$ denotes the data after $x_{ij}$ standardization treatment. $min\left(x_j\right)$ is the minimum value of the jth organizational toughness, and $max\left(x_j\right)$ is the maximum value of the jth organizational toughness.

For positive indicators of growth:

$$x_{ij}^{\prime }={\displaystyle \frac{x_{ij}-min\left(x_j\right)}{max\left(x_j\right)-min\left(x_j\right)}}$$

(2)

For negative indicator volatility:

$$x_{ij}^{\prime }={\displaystyle \frac{max\left(x_j\right)-x_{ij}}{max\left(x_j\right)-min\left(x_j\right)}}$$

(3)

(2)

Calculation of the weight of each indicator:

$$P_{ij}={\displaystyle \frac{x_{ij}^{\prime }}{{{\displaystyle \sum }}_{i=1}^n{x}_{ij}^{\prime }}}$$

(4)

(3)

Calculate information entropy:

$$e_j=-k{\displaystyle \sum _{i=1}^n(p_{ij}\ast ln{p}_{ij})},\hspace{1em}k={\displaystyle \frac{1}{ln(n)}}$$

(5)

(4)

Calculation of weights, $w_j$ denoting the weight of the jth organizational resilience element, which is a reflection of the level of importance of that element, and the greater its value, the greater the impact on organizational resilience.

$$w_j={\displaystyle \frac{1-e_j}{{\displaystyle \sum _{j=1}^n{d}_j}}}$$

(6)

(5)

Calculating the composite score:

$$Res={\displaystyle \sum _{j=1}^n(P_{ij}\ast w_j)}$$

(7)

3.2.2. Independent Variable

The explanatory variable is the digitization of manufacturing. At present, there is no uniform standard for measuring the digitization level; with reference to the research of some scholars, this paper adopts the input-output table in the ADB-MRIO database. Regarding the definition of digitization, drawing on Yu [54], the 35 sectors of the input-output table are screened, and finally the level of the two sectors of the manufacturing industry of computer, electronic, and optical equipment (c14) and the telecommunication equipment and services industry (c27) are selected as proxies for measuring the degree of the level of digitization of the manufacturing industry (c3–c16). The direct and full consumption coefficients are calculated. The direct consumption coefficient refers to the direct consumption amount of other industries required by an industry to produce one unit of product, and is calculated as follows:

$$a_{ij}=S_{ij}/Y_j$$

(8)

In Equation (8), $Y_j$ represents the inputs of all industries in manufacturing $j$, $S_{ij}$ represents the inputs of digital industry $i$ in manufacturing $j$, and $a_{ij}$ represents the level of digitization of manufacturing inputs as measured by the direct consumption coefficient. In addition to direct consumption, each industry also requires indirect consumption in the production process, therefore, the full consumption coefficient is further calculated with the formula:

$$b_{ij}=a_{ij}+{\displaystyle {\sum }_{k=1}^n{a}_{ik}{a}_{kj}}+{\displaystyle {\sum }_{k=1}^m{\displaystyle {\sum }_{k=1}^n{a}_{is}{a}_{sk}{a}_{kj}}}+\cdots$$

(9)

In Equation (9), $b_{ij}$ denotes the level of digitization of manufacturing inputs as measured by the full consumption factor. ${\displaystyle {\sum }_{k=1}^n{a}_{ik}{a}_{kj}}$ is the first indirect consumption of digital industry by manufacturing industry $j$. ${\displaystyle {\sum }_{k=1}^m{\displaystyle {\sum }_{k=1}^n{a}_{is}{a}_{sk}}}{a}_{kj}$ denotes the second indirect consumption of digital industry $i$ by manufacturing industry $j$, and so on adding up to the nth time to obtain the full consumption scenario.

Finally, drawing on Zhang [55], the level of horizontal digitization at the firm level is further estimated according to $D{R}_{ijt}^c=(pc{a}_{it}/{\overline{pca}}_{jt})D{R}_{jt}^c$, where $pc{a}_{it}$ is the firm’s level of capital stock per capita and ${\overline{pca}}_{jt}$ is the average of capital stock per capita across industries.

3.2.3. Intermediary Variable

Technological innovation (Patent): This is measured by taking the natural logarithm of the number of invention patent grants plus one [56].

3.2.4. Moderator Variable

Financial redundancy (FS): This is measured by the ratio of a company’s quick assets to total liabilities. The ratio reflects the level of financial resources at the disposal of the enterprise [57].

3.2.5. Control Variables

In this paper, we select the following control variables: (i) age of the firm (age) [58]; (ii) book-to-market ratio (bm) [59]; (iii) return on net assets (roe) [60]; (iv) gearing ratio (lev) [61]; (v) R&D intensity (rd) [62], (vi) shareholding concentration (top10) [63].

3.3. Data Sources

The data used for variables in this paper mainly come from two groups. One is the ADB-MRIO database, which provides input-output tables from 2007 to 2022. Based on this, the level of digital input in the manufacturing industry can be measured. The other is the CSMAR database, which is used to measure organizational resilience data and a series of control variables. Following the practice of Qin [64], according to enterprise names and stock codes, the Chinese National Economic Industry Classification Standard (GB/T4754-2017) [65] and the ADB database are sorted and matched correspondingly, and the initial samples are processed as follows: ① delete abnormal samples; ② delete ST and PT enterprise samples; ③ perform 1% winsorization on both sides for all continuous variables. Finally, a total of 23,479 sample observations at the enterprise level from 2007 to 2022 are obtained. The descriptive statistics of major variables are shown in

Table 1. Descriptive statistics of the main variables.

Variables	N	Mean	SD	Min	Max	p50
Res	23,479	0.8567	0.1296	0.0568	0.9733	0.8931
Dig	23,479	0.1687	0.2479	0.0009	1.5734	0.0808
roe	23,479	0.0515	0.1712	−1.0517	0.3970	0.0678
bm	23,479	0.5930	0.2344	0.1141	1.1390	0.5913
lev	23,479	0.4115	0.2052	0.0529	0.9669	0.4026
rd	23,479	0.0401	0.0365	0.0000	0.2049	0.0350
age	23,479	1.9528	0.9198	0.0000	3.2958	2.1972
top10	23,479	0.4339	0.1967	0.1314	0.9188	0.3942

. The VIF test results show that the maximum value of the variable is 1.37, which is far below the critical value of 10 for multicollinearity. Therefore, it can be considered that there is no multicollinearity problem among variables.

4. Empirical Results and Analysis

4.1. Benchmark Regression Results

Table 2. Benchmark model regression results.

	(1)	(2)	(3)	(4)
	Res	Res	Res	Res
Dig	0.0306 ***	0.0197 ***	0.0687 ***	0.0270 ***
	(0.00350)	(0.00361)	(0.00968)	(0.00753)
control variable	NO	YES	YES	YES
Constant	0.852 ***	0.827 ***	0.816 ***	0.752 ***
	(0.00103)	(0.00440)	(0.00733)	(0.00610)
individual effect	NO	NO	YES	YES
time effect	NO	NO	NO	YES
N	23,479	23,479	23,479	23,479
R2	0.003	0.052	0.117	0.448

Note: *** indicate significant 1% levels. Robust standard errors clustered to the firm level are in parentheses.

presents the results of the benchmark model regressions. Columns (1) through (4) report the results of regressions that include only explanatory variables, plus control variables, further control for individual effects, and also control for year and firm double fixed effects, respectively. The results indicate that firms’ organizational resilience becomes more resilient as the level of digitization increases. The results in column (4) show that the coefficient of Dig or digitization level is 0.027 and significant at the 1% level after adding control variables and double fixed effects, i.e., for every 1% increase in the digitization level of manufacturing industry, organizational toughness increases by 0.027%, indicating that digitization of manufacturing industry inputs has a significant and positive impact on organizational toughness, and hypothesis H1 is verified.

4.2. Robustness Tests

4.2.1. Substitution of Explanatory Variables

The level of digitization was measured using the direct consumption coefficient method. Substituting the data into the model, the results are shown in column (1) of

Table 3. Robustness test results.

	(1)	(2)	(3)
	Res	Res	Res
	Substitution of explanatory variables	Deletion of special years	Temporal clustering
Dig	0.0182 ***	0.0276 ***	0.0270 ***
	(0.00542)	(0.00756)	(0.00822)
control variable	YES	YES	YES
Constant	0.752 ***	0.773 ***	0.812 ***
	(0.00611)	(0.00561)	(0.0299)
individual effect	YES	YES	YES
time effect	YES	YES	YES
N	23,479	21,922	23,479
R2	0.448	0.231	0.543

Note: *** indicate significant 1% levels. Robust standard errors clustered to the firm level are in parentheses.

. The core explanatory variable remains significantly positive at the 1% level, indicating that the benchmark results are relatively robust.

4.2.2. Impact of the Financial Crisis

Considering the impact of the financial crisis on business resilience during the sample period, we therefore, refer to Kang [66], who restricts the sample year to 2009 and beyond, and the results are relatively robust.

4.2.3. Further Consideration of the Time Factor

Drawing on Zhu [67], a cluster analysis in the time dimension is conducted in the baseline regression to minimize the potential interference that time trends may have on the results of the study, and the regression estimation results are shown in column (3) of Table 3, where the conclusions still hold.

4.2.4. Placebo Testing

The correlation between manufacturing digitization and organizational resilience in the above test results may also just be a natural reflection of the time trend or due to other chance factors. In order to rule out this possibility, this paper refers to Zhang [68], who conducted a placebo test by randomly constructing a manufacturing digitization indicator. The specific operational steps are: first, we randomly generate a dummy variable for manufacturing digitization; second, we replace the manufacturing digitization variable in the original model with this forged digitization variable and re-substitute the sample data into the original model for the regression analysis; lastly, we repeat the above steps 1000 times to guarantee the robustness of the results.

The distribution of the regression coefficients in Figure 1 shows that they exhibit a symmetrical distribution pattern around the value of 0. There is a significant difference in the mean value of these estimated coefficients compared to the coefficient of 0.027 in column (2) of Table 2 in the main test. From the distribution of p-values in Figure 2, we find that for the majority of random samples, the p-values of the regression coefficients are larger than 0.1, implying that these coefficients are not statistically significant. The results support the conclusion in the main test that the relationship between digitization and organizational resilience in the manufacturing industry is real and to a certain extent excludes the interference of time variation and other potential chance factors.

4.2.5. Endogenous Testing

Considering the possible endogeneity problems such as omitted variables and reverse causality in the previous study, the instrumental variable method was employed to conduct a 2SLS test. Drawing on Hao [69], the coefficient of full consumption of digitization inputs in Japan during the same period was chosen as an instrumental variable for the digitization level of Chinese firms. The reasons for selecting this instrumental variable are as follows, the situation of Japan and China during the sample study period is relatively similar, Japan started earlier in industrial informatization and digitization construction, and during the same period, China also began to promote industrial informatization and digitization construction, and Japan’s industry is less affected by China’s economic development, which will not directly affect China’s market, and satisfy the condition of exogeneity of instrumental variables. The regression results of instrumental variables are shown in

Table 4. Regression results of endogenous tests.

	(1)	(2)
	First Dig	Second Res
Dig_japan	1.1797 ***
	(0.01942)
Dig		0.0297 ***
		(0.00778)
control variable	YES	YES
Kleibergen–Paap rk LM	150.639 ***
Kleibergen–Paap rk Wald F	3689.435
individual effect	YES	YES
time effect	YES	YES
N	23,343	23,343
R2	0.928	0.448

Note: *** indicate significant 1% levels. Robust standard errors clustered to the firm level are in parentheses.

. The regression coefficient of the instrumental variable Dig_japan in the first stage is significantly positive, and the Kleibergen–Paap rk LM statistic is significant at the 1% level, which rejects the original hypothesis of “insufficient identification of instrumental variables”, and the Kleibergen–Paap rk Wald F statistic is much higher than the maximum value (16%) of the instrumental variable. The Kleibergen–Paap rk Wald F-statistic is much higher than the critical value of the maximum instrumental variables (16.38), which rejects the hypothesis of “the existence of weak instrumental variables”, and fully explains the reasonableness of the selection of instrumental variables. The regression coefficients of the core variables in the second stage are also significantly positive, which again verifies the robustness of the core conclusions of this paper.

4.3. Heterogeneity Tests

4.3.1. Heterogeneity Test Based on the Nature of Property Rights

Chinese enterprises are characterized by more obvious differential attributes such as the nature of property rights. Enterprises with different attributes may produce different corporate behaviors under the wave of digitization, which in turn have different impacts on organizational resilience. Thus, the sample is divided into state-owned enterprises and non-state enterprises, and the regression results are shown in

Table 5. Results of heterogeneity test (Nature of business and size of business).

	(1)	(2)	(3)	(4)
	Res	Res	Res	Res
	government-owned	non-municipal	broad scale	limited scale
Dig	0.0109	0.0330 ***	0.0133	0.0401 ***
	(0.00884)	(0.00969)	(0.00962)	(0.0142)
control variable	YES	YES	YES	YES
Constant	0.772 ***	0.723 ***	0.774 ***	0.736 ***
	(0.00827)	(0.0111)	(0.00723)	(0.0111)
individual effect	YES	YES	YES	YES
time effect	YES	YES	YES	YES
N	7845	15,634	11,737	11,742
R2	0.736	0.329	0.686	0.342

Note: *** indicate significant 1% levels. Robust standard errors clustered to the firm level are in parentheses.

columns (1)–(2). The results suggest that digitalization of manufacturing has a greater impact on organizational resilience in non-SOEs. The possible reasons for this are: (1) SOEs are subject to more government intervention, which restricts their flexibility and innovativeness in input digitization, whereas non-SOEs are subject to relatively fewer policy and regulatory constraints, which provides more room for them to improve their input digitization level; and (2) SOEs, due to their natural policy tilts, etc., do not have a strong willingness to acquire core competitiveness by improving their digitization level and are not strong. Non-SOEs, on the other hand, tend to face greater competitive pressures in the marketplace, and improving digitalization provides new growth points and competitive advantages for non-state-owned enterprises, helping to improve their organizational resilience.

4.3.2. Heterogeneity Test Based on Firm Size

The role of enterprise manufacturing digitalization on organizational resilience is influenced by enterprise size. Referring to Liu et al. [70], we calculate the median of the natural logarithm of the asset size of the sample enterprises, and set the enterprises with the corresponding value higher than the median as large enterprises, and the rest as SMEs. The regression results are shown in Table 5 columns (3)–(4), which shows that input digitization has a significant positive effect on the organizational toughness of SMEs. Organizational toughness of large enterprises is not significant; the reason may be that large enterprises themselves have a strong ability to bear risks when facing risks, so their impact on organizational resilience is relatively small. While SMEs are not as capable of bearing risks as large enterprises, their business forms are more flexible and they can quickly adapt to changes in the external environment. Driven by digitalization, enterprises can quickly respond to market changes, thereby improving their ability to bear risks and enhancing organizational resilience.

4.3.3. Heterogeneity Test Based on Factor Intensity

Referring to Cong [71], the ISIC codes are matched with ADB-MRIO codes to classify the firms according to the factor intensity characteristics of the industry in which they are located, and the regression results are shown in

Table 6. Results of heterogeneity test (sector).

	(1)	(2)	(3)
	Res	Res	Res
	labor-intensive	capital-intensive	technologically intensive
Dig	0.0758	0.0324	0.0279 ***
	(0.0476)	(0.0409)	(0.00775)
control variable	YES	YES	YES
Constant	0.735 ***	0.760 ***	0.751 ***
	(0.0148)	(0.00875)	(0.00929)
individual effect	YES	YES	YES
time effect	YES	YES	YES
N	2764	7078	13,637
R2	0.511	0.493	0.415

Note: *** indicate significant 1% levels. Robust standard errors clustered to the firm level are in parentheses.

columns (1)–(3). The results show that input digitization positively affects organizational toughness in technology-intensive industries, while it is not significant for labor-intensive and asset-intensive industries, possibly because the core advantage of technology-intensive industries depends on the technological content of the production process, and thus an increase in digital factor inputs directly increases the production inputs of the firms within the industry, improves the efficiency of research and development (R&D), and optimizes the performance of the products, thus enhancing organizational toughness. Labor-intensive industries may have limitations in the application of technology and innovation. The production process in these industries often relies on a large amount of human labor, and investing in digitization may require more time and resources to train employees and change work processes, so the short-term effect on organizational resilience may not be significant. Asset-intensive industries usually require large investments in fixed assets, such as machinery and equipment. Commitment to digitization in these industries may be constrained by existing assets, as the introduction of new technology may require upgrading or replacing existing equipment, which requires larger capital investment and time costs.

4.4. Mechanism Tests

4.4.1. Mediating Effects of Technological Innovation

The results of the mediation mechanism test that manufacturing digitalization promotes the organizational resilience of firms by enhancing their innovation capacity and thus the organizational resilience of firms are shown in

Table 7. Mechanism test results.

	(1)	(2)	(3)
	Patent	Res	Res
Dig	0.279 ***	0.0264 ***	0.0136 *
	(0.0786)	(0.00750)	(0.00800)
Patent		0.00206 **
		(0.000961)
c.Dig#c.FS			0.00784 ***
			(0.00202)
FS			−0.00947 ***
			(0.00112)
control variable	YES	YES	YES
Constant	−0.297 ***	0.752 ***	0.791 ***
	(0.0725)	(0.00611)	(0.00773)
individual effect	YES	YES	YES
time effect	YES	YES	YES
Sobel Z	11.82 ***
Bootstrap_bs_1	[0.0053, 0.0074]
N	23,479	23,479	23,479
R2	0.448	0.274	0.448

Note: *, **, and *** indicate significant at the 10%, 5%, and 1% levels, respectively. Robust standard errors clustered to the firm level are in parentheses.

columns (1)–(2). Among them, column (1) shows that the coefficient of the impact of digitalization on technological innovation is significantly positive, indicating that the enterprise’s investment in digitalization effectively enhances the enterprise’s innovation capability, and the coefficient of Patent in column (2) is significantly positive, indicating that the enterprise’s innovation capability significantly and positively affects the enterprise’s organizational toughness. Further, using the Sobel and Bootstrap tests, the results show that the Sobel Z value is 11.82 and is significant at the 1% level, and the confidence interval of Bootstrap_bs1 does not contain 0 after 1000 times of randomization, which further proves that the mediating effect of enterprise technological innovation is significant, and hypothesis H2 is verified.

4.4.2. Moderating Effects of Financial Redundancy

The regression results of the moderating effect of financial redundancy in the relationship between manufacturing digitization and organizational resilience are shown in column (3) of Table 7. The interaction term shows that the coefficient of the interaction term between manufacturing digitization and financial redundancy is positive and significant at the 1% level. In summary, it can be seen that financial redundancy has a significant strengthening effect on the relationship between manufacturing digitization and organizational resilience, there is a significant positive moderating effect, and hypothesis H2 is verified. As an important reserve resource of the enterprise, financial redundancy can provide sufficient resource support for the enterprise’s innovation activities based on digitalization and the integration and upgrading of digital platforms, which is conducive to enhancing organizational resilience. Therefore, the more financial redundancy, the stronger the positive correlation between manufacturing digitalization and organizational resilience, and hypothesis H3 is verified.

5. Conclusions and Recommendations

Based on theoretical analysis, this paper investigates the impact of manufacturing digitization on organizational resilience and its mechanism of action by using microdata at the firm level of manufacturing industries after the merger of the ADB-MRIO database and the Cathay Pacific database from 2007–2022. It is found that: (1) Manufacturing digitalization inputs can significantly contribute to the enhancement of organizational resilience. (2) Technological innovation plays a partial mediating role in the process of manufacturing input digitalization affecting organizational toughness, specifically, manufacturing digitalization can promote organizational toughness through enhanced technological innovation. (3) The impact of manufacturing digitalization on the organizational toughness of enterprises in different enterprise sizes, different factor intensity industries between the existence of a certain degree of heterogeneity, non-state-owned enterprises, small-scale enterprises, and technology-intensive industries to enhance the role of organizational toughness is more significant. Based on the above conclusions, this paper draws the following policy implications:

Firstly, the government should vigorously promote enterprises to enhance their digital capabilities, strengthen guidance on digital technology research and development and innovation, and optimize the allocation of enterprise innovation resources and decision-making processes. This will facilitate the deep integration of the digital economy with enterprise innovation activities, enabling enterprises to fully seize the innovation opportunities brought about by digital technology. Additionally, the government should encourage enterprises to intensify technological innovation, assist them in accurately identifying risks and taking countermeasures in uncertain general environments, and enhance their ability to handle crises, thereby boosting resilience. The government should also increase the attraction and training of high-end manufacturing talents to stimulate the vitality of the main body of innovation. Broadening the financing channels for digital technology and providing financial support for key and difficult projects is essential. By improving enterprise production efficiency, enhancing industry competitiveness, and reducing the cost of enterprise input digitization through these measures, the adverse impact of external environmental changes on enterprises can be mitigated, laying the foundation for the cultivation of organizational resilience.

Secondly, in the face of the intricate external environment, resisting the impact of external uncertainty events and enhancing the resilience of its own development is an inevitable requirement for China’s manufacturing industry to realize the transformation from a manufacturing power to a manufacturing powerhouse. Therefore, it is crucial to strengthen the construction of organizational resilience and continuously improve the organization’s adaptive and responsive ability to change. In the current situation of increased environmental uncertainty, the future stable development of enterprises needs to rely more on their strong organizational toughness. Designing a flexible organizational structure can facilitate rapid decision-making and resource reconfiguration. Flat management and the establishment of cross-departmental teams can improve the efficiency of information flow and enhance the organization’s responsiveness to external changes. Establishing a sound risk-assessment and management mechanism to regularly evaluate potential market, technological, legal, and environmental risks and develop corresponding response strategies is also essential. To enhance organizational resilience, it is important to smooth the development path of input digitization. This includes vigorously developing digital core technologies in manufacturing, enhancing the intermediary role of technological innovation, and thereby enhancing organizational resilience.

Thirdly, the government should pay attention to the heterogeneous impact of input digitization on organizational resilience across different manufacturing industries. It should set targets at varying levels based on the development characteristics of each industry. For labor-intensive and capital-intensive industries, efforts should be made to integrate digital technology with production practices, gradually forming distinctive, intelligent digital solutions to better enhance organizational resilience. Different measures should also be taken for different types of enterprises. For non-state-owned enterprises, there should be increased investment in scientific research funds, more R&D cooperation opportunities provided, and promotion of the application of digital technology and innovation within these enterprises. This will facilitate the integration of digital technology into their R&D, production, organization, and operations management. Additionally, efforts should be made to leverage the technological and human capital of non-state enterprises and enhance their data mining capabilities. For SMEs, increasing digital investment is more beneficial to improving their market competitiveness. However, their small scale may make it difficult to increase digital construction due to insufficient funds. Therefore, in the future, the government should increase funding for SMEs and encourage them to invest more in digitization through measures such as tax breaks, financial subsidies, and the establishment of more business incubation centers, thereby enhancing their organizational resilience.

6. Limitations and Perspectives

There are some limitations of this study that can be used as a direction for future research. First, due to the limitation of data acquisition, this paper studies a single country, China, and future research could incorporate multiple economies around the world into a unified analytical framework, leading to broader and deeper conclusions. Second, this paper conducted a heterogeneity test based on factor intensity. Future research could categorize them based on other criteria, such as the characteristics of the digital industry and the purpose of the inputs.