The Impact of Blockchain Technology Applications on Enterprise Innovation in the Digital Economy Era: An Empirical Test Based on a PSM-DID Model

Abstract

In the era of digital economy, enterprises need to maximize the acquisition and absorption of innovation knowledge with the help of digital technologies. And blockchain technology, as an emerging digital technology with features such as tamper-evident, distributed storage and accounting, has the potential to help enterprises solve the trust risk in a multi-entity environment, and help them acquire and absorb innovative knowledge to gain innovation advantages. Therefore, this paper deeply explores the impact of blockchain technology applications on enterprise innovation and analyzes the mediating effect of knowledge acquisition and absorptive capacity. Based on the data of Chinese A-share listed companies from 2013 to 2020, this paper conducts an empirical test by applying the PSM-DID method. The results of this study find that blockchain technology applications positively promote enterprise innovation. In terms of impact mechanism, this paper illustrates that blockchain technology applications promote enterprise innovation by improving knowledge acquisition and absorptive capacity. Heterogeneity analysis shows that blockchain technology applications have a more significant promoting effect on innovation in private enterprises. This study expands the theoretical research on the relationship between blockchain technology applications and enterprise innovation, which is of certain significance to the development of enterprise digital innovation under the background of digital economy.

Keywords

blockchain technology applications enterprise innovation knowledge acquisition capacity knowledge absorptive capacity PSM-DID

Introduction

Currently, China’s digital economy has entered the 2.0 era, and the complexity and convergence of technologies required for enterprise innovation and their rapid iterations make it difficult to achieve technological innovation by relying on internal knowledge alone, and more and more enterprises are realizing the importance of building or participating in innovation networks. The key to the impact of digital scenarios on corporate innovation networks is uncertainty, risk, virtualization and dynamism, and the emergence of rapid changes in digital technologies, technological barriers and other issues make innovation networks based on weak connections more vulnerable (Corritore et al., 2020). In this context, knowledge transfer between subjects in innovation networks requires sufficient trust to highlight the role of value co-creation, highlighting the importance of “embedded” relationships (Gomber et al., 2018). The better use of such “network embedding” by firms has a significant positive effect on the efficiency of innovation knowledge acquisition and knowledge absorption capacity (Acemoglu et al., 2018). Therefore, a powerful technology is needed as a tool to improve the trust between innovation agents in order for firms to better “embed” innovation networks and integrate innovation knowledge. Blockchain, as an emerging technology, has changed the way data is stored, retrieved and shared in the digital economy with its anonymity, transparency, security, traceability and efficiency (Böckel et al., 2021), and its rational use has the potential to help companies acquire and absorb more knowledge and information resources, reduce the trust risk faced by innovation agents and promote the success of different innovation agents. The success of different innovation actors is facilitated by reducing the trust risk faced by innovation actors (Korpela et al., 2017). Blockchain technology has characteristics different from those of big data and cloud computing, and its technologies such as immutability, distributed storage and accounting can help enterprises solve the problems of trust deficit such as confirmation and authorization, commodity traceability, and privacy protection, and enable enterprises to establish alliance chains through private chains to promote the exchange of knowledge and information among different enterprises, which may help enterprises gain innovation advantages.

As an emerging digital technology, the research on the application value of blockchain has attracted lively discussions in both academic and practical circles. The existing literature has mainly discussed many aspects of data governance, security regulation, benefit distribution, and customer relationship governance, and all of them have emphasized the characteristics of information fidelity and trustworthiness. In terms of data governance, the characteristics of blockchain technology such as immutability, decentralization and openness and transparency are emphasized to match the need for transparency in the realization of big data value (Lohmer et al., 2020); in terms of security regulation, Schmidt and Wagner (2019) argue that the consensus mechanism and contractual guarantee system of blockchain technology is secure and trustworthy; Ganesh Pillai and Bindroo (2020) believes that the features of blockchain technology such as traceability and intellectual property rights confirmation have a positive impact on customer relationship governance. It can be seen that scholars’ research on the application scenarios of blockchain technology has been extended to new areas such as supply chain, social good, and identity authentication (Queiroz et al., 2019). As the research progresses, the impact of blockchain technology application on innovation gradually draws the attention of scholars, and Saberi et al. (2019) point out that blockchain technology can help build a safe and secure regional innovation system. W. Liu et al. (2022) further analyze the positive impact of blockchain technology on data privacy security and data flow across regions, pointing out that it improves regional innovation through smart contracts and other ways to enhance the efficiency of regional innovation. Tribis et al. (2018) conducted a study in terms of industry chain intelligence and networking, pointing out that blockchain technology has driven the innovation of credit sharing mechanisms in industry chains, which has had a positive impact. It can be seen that while previous studies have affirmed the positive impact of blockchain technology on innovation, mostly based on the impact on the innovation environment and innovation ecology. However, there is a lack of micro-level examination of the impact of blockchain technology on enterprise innovation.

In view of this, this paper attempts to examine the impact of blockchain technology applications on enterprise innovation based on knowledge base theory, innovation network theory, and open innovation theory, etc. The possible research contributions of this paper are the following three points: Firstly, this paper explores the impact effect of blockchain technology applications on enterprise innovation. This paper finds that the emergence of blockchain technology overturns the original trust mechanism, and its decentralization, security, storability and sharing characteristics eliminate the condition of having to trust third-party institutions in data transactions, which accelerates the construction of new digital trust relationships, helps enterprises gain an innovation advantage, and promotes their innovation. Secondly, through the analysis of the mediating effect of knowledge acquisition and absorptive capacity, this paper further reveals the impact mechanism of blockchain technology application on enterprise innovation, providing necessary references for enterprises to effectively acquire external innovation resources and knowledge in the context of the digital economy. In addition, this paper analyzes the differences in the impact of blockchain technology application on corporate innovation by the nature of the firm, and provides the necessary guidance for developing corporate digital innovation strategies. The findings of this paper provide new ideas and theoretical basis for academics to deeply explore the relationship between blockchain technology applications and enterprise innovation.

The follow-up of this paper includes: the second part is theoretical background, the third part is literature review and hypothesis development, the fourth part is research design, the fifth part is research results, the sixth part is further analysis, the seventh part is conclusions and recommendations, and the last part is limitations and directions for future research.

Theoretical Background

Knowledge base theory argues that the firm is essentially a knowledge system and that better use of internal knowledge versus more effective acquisition and assimilation of external knowledge becomes the key to improving the enterprise innovation. The core idea of open innovation theory is that the boundaries of an organization are open, and innovation resources can flow in or out between organizations (Chesbrough & Prencipe, 2008). And innovation network theory argues that innovation networks among firms can serve as effective channels for the transfer of external technological and social knowledge, which affects firms’ innovation. Based on definition of corporate innovation networks, we define corporate innovation networks as the sum of relatively stable, innovation-inspiring, locally rooted, formal or informal relationships between firms and actors in a certain region in an interactive process. Firms need to be “embedded” in innovation networks to gain access to information and knowledge exchange, which is essential to strengthen knowledge sharing mechanisms and facilitate knowledge transfer, and also helps firms to acquire and absorb tacit knowledge needed for innovation, thus positively affecting their innovation (H. Liu et al., 2021). Trust in innovation networks is critical to the acquisition and absorption of tacit knowledge required for innovation, and previous studies have generally classified it under the framework of network relationship embedding as an important factor influencing network relationship embedding (Kraus et al., 2021) and T. Liu and Tang (2020) in their study of the impact of network relationship embedding on firms’ innovation, verified that trust positively influences firms’ innovation by positively acting on innovation knowledge acquisition and innovation knowledge transformation uptake to promote firms’ technological innovation.

The booming development of digital technology is driving the digital transformation of enterprises, and the information era is gradually replaced by the digital era, which brings boundlessness and uncertainty that affects the development of enterprise innovation networks (Cao et al., 2021). The most significant change is that consumers have become one of the innovation network subjects, contributing a wider range of innovation elements. Secondly, weak connections exist among enterprises in the innovative network, leading to a looser network structure and continuous expansion of network scale and heterogeneity. Thirdly, the traditional network governance model is challenged and the risks of network governance are further expanded. In this situation, enterprises need a more efficient tool to help establish trust among innovation agents, effectively solve the problem of trust deficit caused by loose network structure, and enable enterprises to better embed in the innovation network to acquire innovation knowledge (Pan et al., 2019). As a new generation of digital technology, blockchain technology, with its unique decentralization, sharing, security and storage, can effectively reduce the trust barrier among innovation agents, allowing enterprises in innovation networks to quickly establish trust at a lower cost and guarantee the stability of trust between the two parties. Therefore, based on knowledge base theory, innovation network theory, and open innovation theory, this paper explores the impact effect of blockchain technology on enterprise innovation. It also delves into how blockchain technology can help enterprises better obtain innovative knowledge and timely transform and absorb knowledge, thus promoting enterprise innovation.

Literature Review and Hypothesis Development

Blockchain Technology Applications and Enterprise Innovation

In recent years, with the rapid development of digital technology, scholars have begun to focus on the importance of digital technology in the field of business management and innovation, and to explore in depth how digital technology applications can optimize business innovation (Clohessy & Acton, 2019). As an emerging digital technology, blockchain technology with its distributed ledger technology, smart contracts and consensus mechanism can effectively strengthen trust among innovation agents and help enterprises to better embed innovation networks and achieve improved innovation. Trust not only enhances the willingness of data owners to share, but also closes the relationship between both innovation subjects and facilitates knowledge transfer. Blockchain technology has the characteristics of peer-to-peer transmission and asymmetric encryption, which not only guarantees the security of information transmission but also reduces the cost of information interaction, which is conducive to enhancing the strength of trust among innovation subjects, which helps form a free and open atmosphere and enhances the willingness of enterprises to share knowledge. Much of the innovation knowledge required by enterprises in innovation networks is tacit knowledge (Fosso Wamba et al., 2020), such as key know-how or experience, etc. The application of blockchain technology facilitates the exchange and transmission of such innovation knowledge and enhances the scope and speed of acquisition and absorption of enterprise innovation knowledge.

Distributed ledger technology guarantees the security of enterprise innovation knowledge. Enterprises can rely on the asymmetric encryption and authorization technologies of the distributed ledger of blockchain to ensure the security of enterprise innovation knowledge. Blockchain technology records the information of all parties in encrypted form, which strongly guarantees the “clarity of ownership,”“non-tamper ability,” and “privacy security” of enterprise innovation knowledge (Yang et al., 2020), prevents the leakage of confidential information, effectively alleviates the defects of weak data security control, insufficient network security protection technology, the vulnerability to attack of information concentration in the cloud, and the risk of leakage of sensitive data and personal privacy, and provides a more credible big account. The risk of leakage of sensitive data and personal privacy is high, and it provides a more credible big data environment.

Smart contract technology and consensus mechanism help enterprises break through the barriers of information sharing between different organizations and establish collaborative rules. Smart contracts allow trusted transactions without a third party, which are traceable and irreversible. When innovative data are synchronized on the chain, platform sharing, system cascading, and cross-domain alliance information interaction can be realized, and the credit level will be improved with instant information sharing (Westerkamp et al., 2018). At the same time, the multiple consensus mechanisms proposed by blockchain technology are applicable to different application scenarios, enabling enterprises to strike a balance between knowledge sharing efficiency and security.

In summary, the application of blockchain technology effectively strengthens the trust among innovation subjects, helps enterprises better embed in the innovation network, improves the efficiency of innovation knowledge acquisition and absorption, and enhances the innovation of enterprises, which is basically reflected in the following two points: First, the smart contract and distributed ledger of blockchain can help enterprises obtain the advantages of information aggregation, clear ownership and resource sharing, and help enterprises better embed in the innovation network It can facilitate the acquisition and absorption of innovation knowledge. Secondly, enterprises can use blockchain to establish information sharing rules, clarify the division of labor of each subject in cooperation, automatically connect with blockchain through smart contracts in the way of different nodes, realize the synchronization of the whole chain of information on the chain, break the information barriers of different subjects, promote cross-platform, cross-system and cross-union information interaction, and provide infrastructure for data collection, sharing and protection in the process of enterprise science and technology innovation. It provides infrastructure for data collection, sharing and protection in the process of enterprise science and technology innovation. Therefore, the following hypothesis is proposed in this study:

H1: Blockchain technology applications positively affect enterprise innovation.

The Mediating Role of Knowledge Acquisition Capability

Previous studies have shown that in innovation networks, mutual trust between innovation subjects can improve the interaction effect of both, the higher the trust degree of both sides of the transaction the lower the transaction risk, and the establishment of trust helps inter-enterprise knowledge sharing and facilitates the formation of enterprise knowledge acquisition and transformation of scientific and technological achievements. Transaction cost theory suggests that under the condition of limited information and computing power, mutual trust between innovation subjects reduces the transaction cost of interaction. There are insurmountable information barriers in the interaction of enterprises, and the information disadvantaged parties among innovation subjects do not have the ability to screen the authenticity of information; even so, profit-seeking finite rational economic people still have a huge transaction drive. In order to reduce the risk caused by the information barrier, technology supply and demand parties will resort to the power of third parties such as technology intermediaries. Blockchain technology, with its distributed ledger technology, smart contract technology and consensus mechanism, provides a more efficient channel for credit information formation and sharing, eliminating the need for technical intermediaries and enabling high transparency and low transaction costs for information interaction on innovative networks, thus breaking the information asymmetry dilemma (Jin & Shao, 2022). Blockchains are considered “trustless” because they eliminate the condition of having to trust third-party institutions in conducting data transactions. Specifically, blockchain’s peer-to-peer transfer technology helps companies to share innovation resources peer-to-peer, without any technical intermediary during the transfer process, and to broadcast the results across all nodes once they have been shared and confirmed. This helps reduce the security risk of the innovation network, thus reducing the transaction costs caused by opportunistic behavior including the intervention of multiple parties and promoting the stability of the entire integrated innovation network. The decentralization and de-trust mechanism of blockchain technology can solve the trust risk in the multi-subject environment, which can provide enterprises with more opportunities and less costs to obtain high quality and wide range of knowledge resources, and accelerate the speed of knowledge transfer, improve the knowledge acquisition ability of enterprises, and further promote the improvement of enterprise innovation.

On the one hand, blockchain technology application promotes enterprises to conduct efficient external learning and knowledge mining, empowers enterprises to realize interactive learning of internal cognition and external environment, quickly connects internal and external information sources, and extends the breadth and depth of knowledge acquisition. In this process, it continuously absorbs, activates and updates knowledge, thus helping enterprises to tap diversified resources. On the other hand, blockchain technology application helps enterprises to accurately acquire and absorb information related to customer needs and customer psychology (Corritore et al., 2020), and actively explore new technology and product knowledge for maximum satisfaction and beyond customer experience. Under the digitalization scenario, the most significant change of enterprise innovation network is that “consumers” become one of the subjects of innovation network. Blockchain application helps enterprises to grasp consumers’ needs more precisely and clarify the shortcomings of current production process and the direction of technological innovation. It can be seen that blockchain technology has the characteristics of large scale, diversification and high speed for knowledge acquisition, which can help enterprises to explore diversified innovation resources, grasp the correct direction of research and development (R&D), greatly reduce the risk of commercialization failure of R&D results, and continuously acquire and update innovation knowledge to better meet customer needs. Therefore, blockchain technology application improves enterprise knowledge acquisition capability, which in turn promotes enterprise innovation. Therefore, the following hypothesis is proposed in this study:

H2: Knowledge acquisition capability mediates the relationship between blockchain technology application and enterprise innovation.

The Mediating Role of Knowledge Absorption Capacity

Blockchain technology application also effectively enhances the knowledge absorption capacity of enterprises, thus improving their innovation. While accelerating the update of the internal knowledge base of enterprises, blockchain technology application also promotes the collision, fusion and transformation of heterogeneous knowledge absorption. For enterprises, innovative knowledge obtained from outside needs to be organically embedded and transformed according to their own characteristics and integrated with internal technologies before it can be used for enterprise innovation, that is, enterprises need to have the ability to identify, absorb and use external innovative knowledge. Casado-Vara et al. (2018) systematically elaborated the key role played by modern information technology in developing and maintaining the absorptive capacity of enterprises, pointing out that counting blockchain the rapid convergence and proliferation of digital technologies such as blockchain provide important opportunities for firms to improve their absorptive capacity. Blockchain’s distributed ledger technology includes distributed storage and distributed accounting technology, which effectively improves the absorptive capacity and information discernment of enterprises. The open innovation network contains a lot of tacit knowledge, especially forward-looking scientific knowledge, which is difficult to understand, which makes the receiver of innovation knowledge prone to misunderstanding and a large degree of information leakage when absorbing and applying such information. Blockchain technology pools the global storage nodes and builds a huge global unified and globally shared storage pool, and distributes the load data of the storage pool to the nodes of each enterprise, which makes it more efficient to load when enterprises need to use and analyze data. Compared with traditional data analysis, enterprises using blockchain technology have shown significant advantages in terms of massive data categorization, analysis speed and processing data scale (L. Qin & Sun, 2024).

In addition, blockchain has distributed storage and distributed accounting technologies, which help enterprises to be more forward-looking in the analysis of the rapidly changing market environment in the era of digital economy in the early stage of research and development, so that the direction of research and development is more targeted at new products or services that meet the structure of consumer demand to improve the competitiveness of products or services. The use of blockchain technology is also profoundly influencing the way companies do business, and business innovators must not only master the knowledge of innovation, but also grasp the direction of R&D to avoid some ineffective innovation. In addition, the R&D innovation of enterprises in open innovation networks is often process innovation, which is generally based on intensive information processing and search, and the re-improvement or combination of existing technologies to obtain new applications of technology or the creation of new products. By using the distributed storage and accounting technology of blockchain, enterprises can realize cross-platform services, which not only improve the ability of extracting innovation results from existing technologies, but also avoid duplicate and ineffective innovation to a certain extent and reduce the resulting sunk costs.

It can be seen that blockchain technology with its unique advantage of storing and analyzing data becomes an irreplaceable helper in enterprise innovation research and development, which can provide the required data analysis and integration for enterprise innovation in a timely manner (Tzabbar et al., 2022) and improve the ability to absorb, digest, transform and utilize innovation knowledge, thus promoting enterprise innovation. Therefore, the following hypothesis is proposed in this study:

H3: Knowledge absorption capacity mediates the relationship between blockchain technology application and enterprise innovation.

The conceptual framework of this study is shown in Figure 1:

Figure 1.

Research framework diagram.

Research Design

Sample Selection and Data Sources

The research sample of this paper is Chinese listed companies in Shanghai and Shenzhen A-shares from 2013 to 2020. The following treatments are made to the data in this paper: (1) Due to the large differences between the capital structure and accounting system of financial industry enterprises and other enterprises, the financial industry enterprises are excluded in this paper. (2) To avoid the impact of enterprises that have fallen into financial crisis and have irreversible deterioration of business conditions, the ST and *ST enterprises in the sample are excluded in this paper. ST stands for special treatment enterprises. If a listed company has an abnormal financial situation, which leads to the risk of delisting of its shares, or if it is difficult for investors to judge the company’s prospects and investors’ investment interests are damaged, the Exchange will implement a risk warning on its shares. Depending on the situation, risk warning is divided into delisting risk warning (*ST) and other risk warning (ST). (3) Enterprises with serious missing key financial data are excluded in this paper. Total corporate assets, total corporate liabilities, corporate net profit, ROTA, ROE, etc. are the key financial data. (4) In order to alleviate the interference of abnormal values to the empirical model, refer to the method proposed by Chen & Kuo (2017), this paper adjusts all continuous variables by tailing them to the upper and lower 1% of observations. The final sample data includes 12,935 observations from 1761 enterprises. The dependent variable (IP) in this paper is obtained from the data of Innovation Patent Study provided by the State Intellectual Property Protection Bureau, China Research Data Service Platform (CNRDS) and the patent grant data included in the China High Technology Industry Statistical Yearbook. Meanwhile, the criteria for dividing the experimental group and the control group in this paper is whether the enterprise has used blockchain technology, which is obtained from the database of Blockchain Investment provided by China Research Data Service Platform (CNRDS), and we have manually collated and checked the data to ensure the accuracy of the information. Other data are mainly from the China Science and Technology Statistical Yearbook, the China Statistical Yearbook and the CSMAR database.

Variable Explanations

Dependent Variable

Enterprise innovation (IP), this paper draws on Wu et al. (2016), the number of patents granted by enterprises is used as a measure of their innovation, and the types of patents in China are divided into invention patents, design patents and utility model patents, and the natural logarithm of the total number of patents granted plus one is used to measure the enterprise innovation.

Independent Variable

$Trea t_{i} \times Pos t_{t}$ , represents the net effect of blockchain application on the innovation of the firm. Treat is a dummy variable that distinguishes the experimental group from the control group. If the enterprise uses blockchain technology, it belongs to the experimental group, and the variable takes the value of 1. If the enterprise does not use blockchain technology, it belongs to the control group, and the variable takes the value of 0. We adopt the Multi-period DID approach, the date of the first announcement of the use of blockchain by a listed company is taken as the year of the policy shock event, before the policy shock is set to 0, after the shock is set to 1.

Control Variables

(1) Firm characteristics at the firm level, including the firm’s total assets (Ass), the firm’s total liabilities (Lia), the firm’s net profit (NP), the proportion of shares held by the firm’s executives (SH), the proportion of shares held by the first largest shareholder (SHF), and other factors that affect the firm’s decision making; (2) the characteristics of the firm’s executive team, which has been noted in the literature to influence the behavior of the firm’s innovation decisions, and here we include the education of the firm’s CEO (Edu) as a control variable; (3) the firm’s financial status and growth capability, including the firm’s return on assets (ROTA), and the firm’s return on net assets (ROE). Table 1 shows the descriptive statistics of the variables included in this paper.

Table 1.

Descriptive Statistics.

Variable name	Specific meanings	Mean	Standard.	Min	Max
IP	Number of patents granted by enterprises (Log processing)	1.6532	1.1473	0.0000	7.5269
Ass	Total assets of the enterprise (Log processing)	22.2337	1.4873	14.9416	31.0359
SH	Shareholding ratio of corporate officers	10.8370	18.6037	0.0000	99.4496
SHF	Shareholding ratio of the first largest shareholder of the enterprise	34.2202	14.9247	0.2900	89.9900
ROTA	Return on corporate assets	0.0536	0.1390	−2.1947	10.6160
ROE	Corporate Return on Net Assets	0.0457	1.2743	−174.89	8.7150
Lia	Total corporate liabilities (Log processing)	21.3615	1.8487	11.6019	31.0466
NP	Corporate net profit (Log processing)	20.72	20.63	17.51	26.48
Edu	Company CEO education	3.5908	0.8750	1.0000	7.0000
Pay	General budget expenditure of the city where the company is located (Log processing)	18.0420	0.4927	16.0374	18.9687

Propensity Score Matching (PSM)

This article conducted an empirical analysis of propensity score matching (PSM), with reference to the methods of Fu et al. (2021). In order to reduce the effect of sample heterogeneity and ensure the accuracy of DID estimation, it is necessary to carry out a balance test on the samples, and the results of the hypothesis of the balance test under nuclear matching are shown in Table 2. Before matching, there are significant differences between the variables, and the t-value is much larger than 2, and the P-value < 5%, which indicates that the differences between the samples in each dimension are very significant. After matching, the differences between the variables decreased, the t-value decreased, and the P-value became non-significant, indicating that after matching, there was no significant difference between the experimental group sample and the control group sample in terms of the matching variables. In addition, the mean values of the covariates change before and after matching, and the absolute values of the standard deviations of the experimental group and the control group become smaller after matching, and both of them are less than 10%, and the sample distributions of the experimental group and the control group are in good consistency, which passes the balanced hypothesis test.

Table 2.

Balance Test Before and After Matching.

		Mean			t-Test
Sample	State	Experimental group	Control group	Standard deviation (%)	t-Value	p-value
Ass	Before matching	21.8394	22.6369	−86.7	−6.43	.000
	After matching	22.7254	22.7267	−0.23	−0.06	.915
HS	Before matching	10.9305	10.8173	58.61	5.35	.026
	After matching	10.7385	10.6719	3.94	0.23	.273
HSF	Before matching	35.0483	33.4594	43.74	8.34	.016
	After matching	35.9372	36.7168	−4.65	−0.02	.305
ROTA	Before matching	0.0540	0.0552	−72.1	−6.16	.031
	After matching	0.0524	0.0523	0.54	0.06	.424
ROE	Before matching	0.04951	0.04185	37.4	5.43	.001
	After matching	0.04409	0.04172	3.84	1.15	.247
Lia	Before matching	22.9284	19.8227	50.33	8.42	.000
	After matching	20.8172	20.9756	−4.74	−0.93	.347
NP	Before matching	21.4857	19.9521	25.14	8.33	.021
	After matching	20.1638	19.4892	1.94	1.24	.273
Edu	Before matching	3.7093	3.4859	62.72	4.25	.008
	After matching	3.5058	3.5794	−2.34	−0.13	.306

The experimental group in this article is enterprises that apply blockchain technology, while the control group is enterprises that do not apply blockchain technology. In order to analyze the impact of blockchain technology application on enterprise innovation, the nearest neighbor matching method in PSM is used to estimate the overall ATT value. The average processing effect (ATT) of the experimental group measured the difference in enterprise innovation between samples with and without blockchain technology. From Table 3, it can be seen that the difference between the matched experimental group and the control group samples narrowed, indicating that the impact before matching was amplified, and PSM corrected this difference. Meanwhile, the matched ATT is positive and the t-test is significant. It can be seen that the use of the PSM method in this paper can match each experimental group sample to a specific control group sample, making the quasi-natural experiment approximately random.

Table 3.

ATT Results Under Kernel Matching.

Variable	Sample	Experimental group	Control group	Differential	Standard error	t-Test
Y	Before matching	1.539046	1.2270362	0.3120098	0.004037229	2.29
	After matching	1.373903	1.3129296	0.0609794	0.003532046	1.48

Model Design

This study aims to evaluate the impact of blockchain technology on firm innovation using a Propensity Score Matching Difference-in-Differences (PSM-DID) model, with reference to the methods of Z. Qin et al. (2023). The experimental group consists of firms that use blockchain technology, while the control group consists of firms that do not use blockchain technology. To address potential selection bias, the study employs the Propensity Score Matching (PSM) method to find a control group that is well-matched to the experimental group in various aspects. After matching, the two groups show similar trends in the matching variables, indicating parallel trends assumption. Additionally, the study takes into account the timing of firms’ adoption of blockchain technology by using the date of blockchain announcement as the policy shock event year, constructing a multi-period DID model that considers the heterogeneity in the timing of blockchain adoption. The baseline regression model (1) is as follows:

Y_{it} = α_{0} + β_{1} Trea t_{i} \times Pos t_{t} + γ Z_{it} + μ_{i} + λ_{t} + ε_{it}

(1)

$Y_{it}$ represents the innovation of the enterprise, $Trea t_{i} \times Pos t_{t}$ is the independent variable, where Treat is a dummy variable that distinguishes the experimental group from the control group. If the enterprise uses blockchain technology, it belongs to the experimental group, and the variable takes the value of 1. If the enterprise does not use blockchain technology, it belongs to the control group, and the variable takes the value of 0. The variable Post represents the policy shock time. We adopt the Multi-period DID approach, the date of the first announcement of the use of blockchain by a listed company is taken as the year of the policy shock event, before the policy shock is set as Post = 0, after the shock is set as Post = 1. $Z_{it}$ represents a series of control variables; $μ_{i}$ represents individual fixed effects; $λ_{t}$ represents a fixed time effect; $ε_{it}$ is a random Error term. The double difference coefficient we are concerned about $β_{1}$ , it’s economic significance can be explained as the actual promoting effect of blockchain technology application on enterprise innovation.

Research Results

Main Effects Test

This article uses the PSM-DID method to test whether blockchain technology has a positive promoting effect on enterprise innovation and systematically evaluates whether there is a significant difference compared to enterprises that do not use blockchain. This article gradually adds control variables and bidirectional fixed effects to the model, and the regression results are shown in Table 4. This paper first constructed panel data covering multiple industries from 2013 to 2020, and then carried out the regression of the double difference model. The result reported in column (1) is that the main variables were simply regressed first, without adding the two-way fixed effect, which can be compared with the later results. The regression results show that the DID coefficient is positively significant at the level of 1%, which proves that the application of blockchain technology positively promotes enterprise innovation. In columns (2) to (3), we used the two-way fixed effect, adding control variables such as corporate growth ability and corporate balance of payments. After adding control variables, the coefficient of influence slightly increases, the coefficient rises to 18.22%, and is significant at the 1% significance level. At the same time, in order to prevent inter group heteroscedasticity and other issues, clustering robust standard error was used to make the results more credible. In order to prevent the influence of variable units on the regression results, this article standardized the main variables with Z-Score to make the regression results more reliable. The double difference regression results show that the difference in innovation achievements between enterprises using blockchain and other enterprises after 2016 is positively correlated with whether or not they use blockchain, and the results are significant. This also proves that the application of blockchain technology has a positive impact on enterprise innovation. In addition, this article further selects all control variables to be lagged for one period and introduces the GMM model for regression, which may avoid endogeneity problems caused by mutual causality, and the results are shown in column (4). We can also find that the application of blockchain technology has a significant positive effect on enterprise innovation.

Table 4.

Main Effects Test Results.

Variables	(1)	(2)	(3)	(4)
DID	0.1637** (0.0236)	0.1754* (0.0271)	0.1822** (0.0356)	0.1837*** (0.0439)
Ass			0.0052** (0.0085)	0.0068* (0.0093)
ROTA			0.0792* (0.1639)	0.0907** (0.1809)
ROE			0.0095 (0.0262)	0.0089 (0.0214)
Edu			0.0032 (0.0064)	0.0028* (0.0057)
Lia			−0.0027* (0.0082)	−0.0015 (0.0055)
NP			0.0003** (0.0049)	0.0002** (0.0060)
SH			−0.0008* (0.0073)	−0.0006 (0.0005)
SHF			0.0003 (0.0009)	0.0004 (0.0006)
Constant	0.4174* (0.2815)	0.2834 (0.0138)	0.1375 (0.0241)	0.1438* (0.0737)
Observations	12,935	12,935	12,935	12,935
R-squared	.0242	.0109	.0192	.0138
Individual	NO	YES	YES	YES
Year	NO	YES	YES	YES

Note. t-Statistics in parentheses: *p < .1. **p < .05. ***p < .01.

Previous studies lacked the use of differences-in-differences method and the use of micro-firm data to empirically analyze the impact of blockchain technology use on firm innovation, which this paper does its best to improve. This paper empirically tests the promotion effect of blockchain technology application on enterprise innovation. Blockchain has the advantages of decentralization, high security, high transparency, traceability, etc., and these features bring many new innovation and development opportunities for enterprises. Enterprises can use blockchain technology to establish smart contracts, realize digital identity authentication, etc., to help them gain an Innovation advantage. Through continuous exploration and innovation, enterprises can utilize blockchain technology to create more innovative opportunities and value. Overall, this paper uses microdata of Chinese enterprises, and the PSM-DID method, to empirically verify that blockchain technology application positively affects enterprise innovation, which also supports our hypothesis.

Effectiveness Analysis of DID Estimation

This section will conduct a series of DID validity tests, including parallel trend tests and placebo tests, to verify the credibility of the empirical results.

Parallel Trend Test

The model setting method we use here draws on the approach of Thiele and Navarro (2014), which states that there is no significant difference in the trend of the experimental group and the control group over time before the policy occurs, but the trend becomes different after the policy occurs. This article examines the trend of changes before and after the policy occurrence point, and Figure 2 reports the analysis results of parallel trend tests in the first 3 years and the following 3 years of the policy occurrence point. From the results, it can be seen that the trend of change between the treatment group and the control group before the application of blockchain technology was consistent, and the interaction coefficient was not significant. After the application of blockchain technology, there was a significant change between the treatment group and the control group, and the development trend of the treatment group showed a significant increase compared to the control group, passing the parallel trend test.

Figure 2.

Parallel trend test.

Placebo Test

This article draws inspiration from Fu et al. (2021) method, which advances the policy occurrence time by 1 to 3 years. This article sets the policy impact time as pre_ 1, pre_ 2, pre_3 .Table 5 reports the corresponding estimated results. According to Table 3, it can be found that the estimation coefficients of the core variables are not significant, so other potential unobservable factors can be excluded from affecting the innovation of the enterprise in this article.

Table 5.

Placebo Test.

Variables	(1)	(2)	(3)
DID	−0.0381 (0.6834)	0.1624 (0.5071)	−0.319 (0.5734)
Control	YES	YES	YES
Observations	12,935	12,935	12,935
R-squared	.0121	.0184	.0075
Individual	YES	YES	YES
Year	YES	YES	YES

Note. t-Statistics in parentheses: *p < .1. **p < .05. ***p < .01.

Robustness Tests

The PSM test has been carried out in the previous article, which has increased the robustness of the empirical results to a certain extent. In addition, this paper also adds macro fiscal factors and urban fixed effects to test the robustness of the empirical results. The specific analysis is as follows.

(1) Add macro fiscal factors. Due to the close relationship between the innovation behavior of enterprises and the external environment (Yang et al., 2017), there are certain differences in the development of different regions in China, which has also become an external factor affecting enterprise innovation behavior. Therefore, we have added the control variable government fiscal expenditure (Pay) here. Local fiscal models have a significant impact on the relationship between government and enterprises and the development of regions, and also affect the choice of innovative behavior by enterprises. In areas with higher government fiscal expenditure, enterprises may face higher government subsidies. Therefore, we have added the fiscal expenditure (Pay) of the city where the enterprise is located. The regression results are shown in Table 6 (1), we found that the results are still positive and significant at the 5% confidence level, which also supports our hypothesis.

(2) Add urban fixed effects. Due to the involvement of multiple companies in our sample, located in different regions and influenced by different external environments, controlling only for time fixed effects and company fixed effects may overlook unobservable factors at the city level. Therefore, we added urban fixed effects to column (2) for analysis. We found that the results are still positively significant, indicating that our results are robust.

Table 6.

Robustness Test.

Variables	(1)	(2)
DID	0.1624** (0.1719)	0.1359* (0.1727)
Observations	12,935	12,935
R-squared	.0265	.0379
Individual	YES	YES
Year	YES	YES
Urban	NO	YES
Control	YES	YES

Note. t-Statistics in parentheses: *p < .1. **p < .05. ***p < .01.

Further Analysis

Mediating Effect Test

In exploring the mediating effect of knowledge acquisition, this study used a three-step approach to test the mediating effect of knowledge acquisition, as modeled by:

Y_{it} = α_{1} + β_{2} Bl O_{it} + γ_{1} Z_{it} + μ_{i} + λ_{t} + μ_{it}

Mediatio n_{it} = α_{2} + β_{3} Bl O_{it} + γ_{2} Z_{it} + μ_{i} + λ_{t} + ω_{it}

Y_{it} = α_{3} + β_{4} Bl O_{it} + β_{5} Mediatio n_{it} + γ_{3} Z_{it} + μ_{i} + λ_{t} + ξ_{it}

Mediatio $n_{it}$ is the mediator variable for this article, $Bl O_{it} = Trea t_{i} \times Pos t_{t}$ , $Z_{it}$ represents a series of control variables; $μ_{i}$ represents individual fixed effects; $λ_{t}$ represents a fixed time effect; $ε_{it}$ is a random Error term.

Mediator variable: Knowledge acquisition capacity (AC), according to Li et al. (2021), the number of patents cited in annual patent applications represents enterprise acquisition capacity. This paper uses the natural logarithm of the number of citations of enterprise patents plus 1 to express the ability to acquire knowledge. knowledge absorption capacity (AB), this paper draws on the practice of Yang et al. (2020) and uses the ratio of annual R&D investment to operating income of sample enterprises.

Table 7 (1) and (2) are estimates of knowledge acquisition capabilities, which report that enterprises use blockchain technology to significantly improve knowledge acquisition capabilities, and model estimates show that enterprises use blockchain technology to improve knowledge acquisition capabilities, thereby promoting the growth of enterprise innovation, further validating the hypothesis 2. In Table 5, (3) and (4) are the estimated results of knowledge absorption, which report that enterprises use blockchain technology to significantly improve knowledge absorption, and the model estimation results show that enterprises use blockchain technology to improve knowledge absorption, thereby promoting the growth of enterprise innovation, further validating hypothesis 3.

Table 7.

Mediating Effect Test.

	(1) AC	(2) $Y_{it}$	(3) AB	(4) $Y_{it}$
Blo	0.3057** (0.0849)	0.2079** (0.0428)	0.0873* (0.0104)	0.1735** (0.0396)
AC		−0.2394*** (0.0537)
AB				0.1940* (0.0454)
Control	YES	YES	YES	YES
Bootstraptest	Z = −4,64, p = .0000		Z = 3.57, p = .0002
Individual	YES	YES	YES	YES
Year	YES	YES	YES	YES
R-squared	.0148	.0129	.0152	.0138
Observations	12,935	12,935	12,935	12,935

Note. t-Statistics in parentheses: *p < .1. **p < .05. ***p < .01.

It can be seen that this paper empirically tests that blockchain technology application can promote enterprise innovation by improving knowledge acquisition and knowledge absorption, through using microdata of Chinese enterprises and three-step approach, empirically analyzes its impact mechanism, which makes up for the shortcomings of previous studies that less empirically analyze the relevant impact mechanism. The application of blockchain technology can help enterprises realize distributed storage and access of knowledge, significantly improve the efficiency of intellectual property protection, and make knowledge information easier to share and utilize. This helps to promote the acquisition and absorption of knowledge by enterprises and further promotes enterprise innovation, which also supports our hypothesis.

Heterogeneity Test

Enterprises innovation behavior is driven by a variety of factors and there is the possibility of heterogeneous behavior in many dimensions (Acemoglu et al., 2018), and discussing heterogeneous factors can help us understand more accurately the impact of using blockchain on corporate innovation behavior, and here we focus on the differences in the nature of firms. The differences in enterprise nature between state-owned enterprises themselves and private enterprises in the Chinese scenario also affect the relationship between the two types of enterprises and the government, and also cause differences in the development model and innovation efficiency of the two types of enterprises (Kopyto et al., 2020), which affects the goals of development of the two types of enterprises and the motivation of enterprise decisions (Mubarak & Petraite, 2020). In this paper, the sample was divided into state-owned enterprises and private enterprises for heterogeneity testing according to the nature of the enterprises to which the sample enterprises belonged, and it was found that the blockchain application had a greater and statistically significant effect on enhancing the innovation of private enterprises, but the effect on state-owned enterprises was not significant. This may be due to the fact that private enterprises are more focused on economic efficiency and have a tougher external business environment than state-owned enterprises, and are more motivated to use blockchain technology to explore the value behind big data. Therefore, the application of blockchain technology may cause the difference in innovation between private and state-owned enterprises. The specific results are shown in Table 8.

Table 8.

Heterogeneity Test Between State-owned and Private Enterprises.

Variables	(1)	(2)
	State-owned enterprises	Private enterprises
DID	0.1173 (0.2891)	0.2326** (0.1253)
Control	YES	YES
Individual	YES	YES
Year	YES	YES
Observations	5,924	7,003
R-squared	.0089	.0199

Note. t-Statistics in parentheses: *p < .1. **p < .05. ***p < .01.

Conclusions and Recommendations

Conclusions

This paper explores the impact effect and influence mechanism of blockchain technology on corporate innovation, tested the actual effect, and conducted heterogeneity analysis by using A-share listed companies from 2013 to 2020 as research objects. It is found that the use of blockchain technology has a positive contribution to corporate innovation. Second, this paper illustrates the impact of blockchain technology on enterprise innovation through improving knowledge acquisition and absorption, and analyzes its intrinsic mechanism. Third, the impact of blockchain technology on firms’ innovation can show heterogeneity depending on the differences in the nature of different firms.

The contribution of this study is mainly reflected in the following aspects: Firstly, this paper expands the research on the impact effect and influence mechanism of blockchain technology application on enterprise innovation, which makes up for the inadequacy of the existing research that pays less attention to the relationship between blockchain technology and enterprise innovation at the micro level. Secondly, this paper integrates innovation network theory, knowledge base theory, open innovation theory and digital technology application, which expands the application scope of traditional theories and injects new perspectives for enterprise digitalization research and enterprise innovation research.

Recommendations

The above findings have the following recommendations for enterprises and the government: First, enterprises should recognize that blockchain technology is an efficient and secure emerging technology, and mastering the use of this technology is beneficial to their own innovation development. It is found that blockchain technology can provide enterprises with the required data analysis capability in a timely manner and improve the ability to digest, convert and utilize innovative knowledge; second, enterprises need to dig deeper into the underlying blockchain protocols and fully assess the security risks. Exploring the combination of blockchain technology and traditional information security technology and integrating their respective advantages can enable blockchain technology to be effectively applied in a wide range of fields; thirdly, the government needs to increase the promotion of emerging technologies, help enterprises crack the technical threshold, prompt more enterprises to join the ranks of using blockchain, give full play to the advantages of China’s blockchain technology and form a synergy to promote enterprise innovation; fourthly, relevant departments need to increase the Fourth, the relevant departments need to increase the legislation on the use of emerging technologies such as blockchain and promote the introduction of standardized norms for emerging technologies, so that the use of blockchain technology can receive comprehensive legal protection.

Limitations and Directions for Future Research

First of all, based on the availability of research data, this paper uses the data of Chinese A-share listed companies, China as a big country in the world, the number of listed companies is large and covers a variety of industries, which is of certain research value for quantitative analysis. However, China’s national conditions are special, and the research on enterprises in other countries may not be universal. In order to explore the wider impact of blockchain technology application on global enterprise innovation, subsequent research can consider using data from other countries in the world. Moreover, the development status of developing and developed countries is different, and it is worthwhile for academics to conduct research in the future to find out whether the impact of blockchain technology application on enterprises in developed countries is different from that of enterprises in developing countries. Secondly, this paper analyzes the mediating effect of innovation knowledge acquisition capacity and absorptive capacity, and there may be other mediating transmission mechanisms, which deserve further in-depth discussion in future research. Future research can also expand the research on the channels of influence of blockchain technology application on enterprise innovation from other perspectives, such as the role mechanism of blockchain technology influencing enterprise innovation can be explored from the aspects of organizational culture, employee skills and market dynamics. Finally, external environmental factors such as environmental dynamics and environmental complexity may modulate the relationship between blockchain technology applications and enterprise innovation, and future research can add relevant moderating variables to the theoretical model, so as to make the theoretical model more complete.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: (1) Graduate Research and Innovation Projects of Jiangsu Province (KYCX22_1388) Research on Innovation Spillover Effect of OFDI by Chinese Digital Economy Enterprises. (2) National Social Science Fund of China (20BJY001) Research on the Development of Strategic Emerging Industries Driven by Domestic Consumption Upgrading.

ORCID iD

Qiu Peiyao

Data Availability Statement

The datasets used during the current study available from the corresponding author on reasonable request.

References

Acemoglu

Akcigit

Alp

Bloom

Kerr

(2018). Innovation, reallocation, and growth. American Economic Review, 108(11), 3450–3491.

Böckel

Nuzum

A. K.

Weissbrod

(2021). Blockchain for the circular economy: Analysis of the research-practice gap. Sustainable Production and Consumption, 25, 525–539.

Cao

Xing

Zhang

(2021). Effect of dual network embedding on the exploitative innovation and exploratory innovation of enterprises-based on the social capital and heterogeneous knowledge. Technology Analysis and Strategic Management, 33, 638–652.

Casado-Vara

Prieto

Corchado

J. M.

(2018). How blockchain improves the supply chain: Case study alimentary supply chain. Procedia Computer Science, 134(2), 393–398.

Chen

P. T.

Kuo

S. C.

(2017). Innovation resistance and strategic implications of enterprise social media websites in Taiwan through knowledge sharing perspective. Technological Forecasting and Social Change, 118(6), 55–69.

Chesbrough

Prencipe

(2008). Networks of innovation and modularity: A dynamic perspective. International Journal of Technology Management, 42(4), 414–425.

Clohessy

Acton

(2019). Investigating the influence of organizational factors on blockchain adoption: An innovation theory perspective. Industrial Management & Data Systems, 119(7), 1457–1491.

Corritore

Goldberg

Srivastava

S. B.

(2020). Duality in diversity: How intrapersonal and interpersonal cultural heterogeneity relate to firm performance. Administrative Science Quarterly, 65(2), 359–394.

Fosso Wamba

Queiroz

M. M.

Trinchera

. (2020). Dynamics between blockchain adoption determinants and supply chain performance: An empirical investigation. International Journal of Production Economics, 229, 107791.

10.

Luo

(2021). Does the low-carbon city policy make a difference? Empirical evidence of the pilot scheme in China with DEA and PSM-DID. Ecological Indicators, 122, 107238.

11.

Ganesh Pillai

Bindroo

. (2020). Supplier cluster characteristics and innovation outcomes. Journal of Business Research, 112, 576–583.

12.

Gomber

Kauffman

R. J.

Parker

Weber

B. W.

(2018). On the Fintech revolution: Interpreting the forces of innovation, disruption, and transformation in financial services. Journal of Management Information Systems, 35, 220–265.

13.

Jin

Shao

Y. F.

(2022). Power-leveraging paradox and firm innovation: The influence of network power, knowledge integration and breakthrough innovation. Industrial Marketing Management, 102, 205–215.

14.

Kopyto

Lechler

von der Gracht

H. A.

Hartmann

(2020). Potentials of blockchain technology in supply chain management: Long-term Judgments of an international expert panel. Technological Forecasting and Social Change, 161, 120330.

15.

Korpela

Hallikas

Dahlberg

(2017). Digital supply chain transformation toward blockchain integration [Conference session]. Proceedings of the 50th Hawaii International Conference on System Sciences.

16.

Kraus

Jones

Kailer

Weinmann

Chaparro-Banegas

Roig-Tierno

(2021). Digital transformation: An overview of the current state of the art of research. Sage Open, 11, 2158.

17.

Cao

Wang

(2021). Organizational mindfulness towards digital transformation as a prerequisite of information processing capability to achieve market agility. Journal of Business Research, 122, 700–712.

18.

Liu

Wang

(2021). Is there any difference in the impact of digital transformation on the quantity and efficiency of enterprise technological innovation? Taking China’s agricultural listed companies as an example. Sustainability, 13, 12972.

19.

Liu

Wang

Jia

Choi

T. M.

(2022). Blockchain announcements and stock value: A technology management perspective. International Journal of Operations & Production Management, 42(5), 713–742.

20.

Liu

Tang

(2020). Open innovation from the perspective of network embedding: Knowledge evolution and development trend. Scientometrics, 124, 1053–1080.

21.

Lohmer

Bugert

Lasch

(2020). Analysis of resilience strategies and ripple effect in blockchain-coordinated supply chains: An agent-based simulation study. International Journal of Production Economics, 228, 107882.

22.

Mubarak

M. F.

Petraite

(2020). Industry 4.0 technologies, digital trust and technological orientation: What matters in open innovation? Technological Forecasting and Social Change, 161, 120332.

23.

Pan

Song

M. L.

Zhang

Zhou

(2019). Innovation network, technological learning and innovation performance of high-tech cluster enterprises. Journal of Knowledge Management, 23, 1729–1746.

24.

Qin

Sun

S. L.

(2024). Knowledge collaboration in global value chains: A comparison of supplier selection between a forerunner and a latecomer. Asia Pacific Journal of Management, 41, 51–79.

25.

Qin

Wang

(2023). Policy assessment of Hubei Province’s dual innovation demonstration base based on PSM-DID method. Statistics and Decision Making, 39(05), 185–188. https://doi.org/10.13546/j.cnki.tjyjc.2023.05.034

26.

Queiroz

M. M.

Telles

Bonilla

S. H.

(2019). Blockchain and supply chain management integration: A systematic review of the literature. Journal of Supply Chain Management, 25(2), 241–254.

27.

Saberi

Kouhizadeh

Sarkis

Shen

(2019). Blockchain technology and its relationships to sustainable supply chain management. International Journal of Production Research, 57(7), 2117–2135.

28.

Schmidt

C. G.

Wagner

S. M.

(2019). Blockchain and supply chain relations: A transaction cost theory perspective. Journal of Purchasing and Supply Management, 25(4), 1–13.

29.

Thiele

T. E.

Navarro

(2014). “Drinking in the dark”(DID) procedures: A model of binge-like ethanol drinking in non-dependent mice. Alcohol, 48(3), 235–241.

30.

Tribis

El Bouchti

Bouayad

(2018). Supply chain management based on blockchain: A systematic mapping study. MATEC Web of Conferences, 200(2), 00020.

31.

Tzabbar

Cirillo

Breschi

(2022). The differential impact of intrafirm collaboration and technological network centrality on employees’ likelihood of leaving the firm. Organization Science, 33, 2250–2273.

32.

Westerkamp

Victor

Kupper

(2018). Blockchain-based supply chain traceability: token recipes model manufacturing processes [Conference session]. Proceedings of the IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData).

33.

Wang

Hong

Piperopoulos

Zhuo

(2016). Internationalization and innovation performance of emerging market enterprises: The role of host-country institutional development. Journal of World Business, 51(2), 251–263.

34.

Yang

Zeng

Zou

Zhao

S. C.

(2020). A study of scientific collaboration network, knowledge diversity and firm innovation. Studies in Science of Science, 39(5), 867–875.

35.

Yang

Fan

Shao

Yang

(2017). Does carbon intensity constraint policy improve industrial green production performance in China? A quasi-DID analysis. Energy Economics, 68(1), 271–282.