Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

RETRACTED: Firm information disclosure environment and R&D investment: Evidence from Internet penetration

Retraction

The PLOS One Editors retract this article [1] because it was identified as one of a series of submissions for which we have concerns about potential manipulation of the publication process, peer review integrity, and authorship. These concerns call into question the validity and provenance of the reported results. We regret that the issues were not identified prior to the article’s publication.

MAK did not agree with the retraction. YW, CL, NL, and RY either did not respond directly or could not be reached.

9 Oct 2025: The PLOS One Editors (2025) Retraction: Firm information disclosure environment and R&D investment: Evidence from Internet penetration. PLOS ONE 20(10): e0333979. https://doi.org/10.1371/journal.pone.0333979 View retraction

Abstract

Guided by the conviction that “Clear waters and green mountains are as good as mountains of gold and silver”, China highly values sustainable economic and social development through innovation and Internet technology. Regression analysis is performed to examine the impact of corporate information disclosure environment proxied by the Internet penetration rate on innovation. Leveraging from the city-level Internet penetration rates data in China from 2003 to 2017, this study gets the following findings: (1) Firms headquartered in cities with high Internet penetration rates tend to be more innovative, i.e. they invest more in research and development. (2) This result is supported by several robustness checks, such as alternative measures of key variables, alternative empirical specifications, and tests to mitigate identification concerns. (3) "financing constraint" and "tolerance of innovation failure" are two channels that influence firms’ innovative endeavors. (4) Additional tests show that Internet penetration rates facilitate a firm’s output efficiency of innovation input, total factor productivity, and human capital environment for innovation. The above conclusions not only enrich the relevant literature on the influencing factors of corporate innovation from the perspective of the firm information disclosure environment but also provide an important reference for further understanding the positive role of macro technology development on social and economic development.

Citation: Wang Y, Li C, Khan MA, Li N, Yuan R (2021) RETRACTED: Firm information disclosure environment and R&D investment: Evidence from Internet penetration. PLoS ONE 16(3): e0247549. https://doi.org/10.1371/journal.pone.0247549

Editor: J E. Trinidad Segovia, University of Almeria, SPAIN

Received: December 12, 2020; Accepted: February 9, 2021; Published: March 18, 2021

Copyright: © 2021 Wang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: This research is funded by the major project of education department of Hebei province: “Research on the path selection and countermeasures for accelerating the transformation and upgrading of state-owned enterprises in Hebei province” (ZD201904). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors declare no conflict of interest.

1. Introduction

"Sustainable development" is the guiding concept of development in the 21st century [1]. The internet, as the leading technology in the 21st century, has been the basis for cultivating various technologies and a strategic cornerstone to ensure the smooth progress of sustainable development. The report of the 19th National Congress of the Communist Party of China has proposed to promote the in-depth integration of the Internet, big data, artificial intelligence, and the real economy, and cultivate new sustainable growth points and form new momentum in areas such as innovation leadership. It also emphasizes the urgency of building an innovative country, to stimulate and protect entrepreneurship, and encourage more social entities to devote themselves to innovation and entrepreneurship. Internet will continue to promote the creation of new products and business formats and give play to the "hardware" role of information infrastructure (National Internet Information Office of the People’s Republic of China: http://www.cac.gov.cn/2020-02/03/c_1582289659655350.htm). So, it is vital to strengthen the ability of reform and innovation, maintain an enterprising spirit, combine actual and creative work, and especially, use the Internet technology and information technology to carry out work. It can be seen that combining the two critical areas of Internet development and corporate innovation and exploring the relationship between the two has important significance.

Internet technology has increasingly become a new driving force for global economic growth. An important feature of the new generation of information technology represented by the Internet is the interconnection of information. It breaks the constraints of time and space and significantly reduces the asymmetry of Information between external investors and companies. With the increase in Internet penetration, investors can quickly obtain and publish information about listed companies with only one device connected to the Internet [2]. For example, the operation of the Shenzhen Stock Exchange’s "Interactive Easy" network platform enables investors to track the business activities of listed companies at any time and keep abreast of company development trends [3]. Besides, the Internet era is an era of innovation and change. From the perspective of the information communication and dissemination functions of the Internet, it is appropriate to use the Internet penetration rate as a proxy for the corporate information disclosure environment. It is hypothesized that Internet penetration rate is valuable because it helps improve corporate information disclosure environment, which, in turn, fosters firm innovation.

Leveraging from the city-level Internet penetration rates data in China from 2003 to 2017, this paper conducts an in-depth discussion on the relationship between Internet penetration and corporate innovation. The main finding are as follows: First, firms headquartered in cities with high Internet penetration rates tend to be more innovative, i.e. they have higher R&D (corporate’s research and development expenditure, Hereinafter referred to as R&D) intensity. This effect is still valid after a series of robustness tests, such as alternatives of key variables, regression model, fixed effects and standard deviation clustering methods; and treatments of endogeneity, both via propensity score matching (PSM) and Heckman two stages model. Secondly, based on the above empirical conclusions, two mechanisms of "financing constraints" and "innovation failure tolerance" have been proposed and proved. Finally, with the increase in the penetration rate of the regional Internet, the transformation rate of innovation achievements and production efficiency faced by firms has been improved. Meanwhile, the high-end element combination required for research and development; that is, the human capital environment has been improved.

This paper provides at least three major contributions to the literature. First, the results in this article contribute to our understanding of the relationship between Internet technology and corporate innovation. Existing literature mainly focuses on the Internetization of corporate entities [4], while this study pays close attention to the information disclosure environment faced by external investors (including shareholders and creditors). Second, it has enriched the literature on the relationship between the information disclosure environment and corporate innovation. Existing research mainly studies the impact of the quality of the information environment on corporate innovation from the perspectives of firm voluntary disclosure and external information intermediary disclosure [5–7]. This paper focus on the impact of the development of macro information technology infrastructure on the information disclosure environment, and then on corporate innovation. Third, our paper is the first to employ the city-level Internet penetration rate, which has substantial geographic variation in firms’ headquarters. This indicator is more precise than the provincial-level one to capture the impact of Internet penetration on R&D intensity [4, 8].

The structure of the other parts of the article is arranged as follows: the second part, literature review and research hypotheses; the third part, research design; the fourth part, the analysis of empirical results; the fifth part, the conclusions.

2. Literature review and research hypothesis

2.1 Literature review

Based on our topic, two types of literature have been reviewed. The details are as follows: First, this paper combs the literature on the relationship between the Internet penetration rate and the corporate information disclosure environment. Next, the literature is reviewed about the relationship between the information disclosure environment and corporate innovation.

The first type of literature: the relationship between Internet penetration rate and corporate information disclosure environment. In recent years, China’s Internet penetration rate has increased rapidly, and it has shown a prosperous development trend. From about 0% in the early 1990s, it has risen to 64.5% in March 2020. The number of Internet users has reached 904 million, and the digital divide keeps shrinking ("The 45th Statistical Report on China’s Internet Development Status" issued by China Internet Network Information Center: http://www.cnnic.net.cn/hlwfzyj/hlwxzbg/hlwtjbg/202004/P020200428596599037028.pdf). With the rapid development of Internet penetration, the role of the Internet in reducing the cost of information search has also been widely confirmed [9, 10].

Existing studies have mainly expounded on the important role of the Internet in improving the corporate information disclosure environment from the perspective of its’ information and communication functions [2]. All market participants attach great importance to the advantages of the Internet in information communication and dissemination. Especially in 2009, the Shenzhen Stock Exchange established an Internet-based "listed companies-investors interactive platform" (referred to as "Interactive Easy" platform). Individual investors get a more convenient opportunity to participate in management interaction. Investors can consult the executives of listed companies with the help of the "Interactive Easy" platform for any questions about the company’s business information. And finally, it improves the accuracy of the information they acquire [11]. At the same time, the opening of the online platform improves the timeliness of information dissemination. The operation of the Internet platform enables investors to track the business activities of listed companies at any time and keep abreast of the company’s development trends. Zhao, Tan [12], Hollenbeck [13] show that the Internet can transmit not only information but also store information. Any company’s credit status, contract performance records, technical level, product quality, reputation, and other information may be published, stored, and disseminated online. The Internet greatly facilitates interpersonal communication, increases the transparency of corporate information, and reduces information imbalance [14]. With the increasing penetration rate of the Internet, the academic community has paid more and more attention to its function in improving the corporate information disclosure environment. Scholars have further researched and found that Internet finance based on the Internet and other information technologies provides new opportunities for solving corporate financing problems. For example, Wang et al. [15] believe that the rapid development of information communication technology, such as mobile Internet, cloud computing, and big data has driven the progress of financial technology and provided technical support for Internet finance. Compared to traditional financial institutions, Internet finance also plays a more positive role in terms of improving financial efficiency, solving asymmetric information, and alleviating the financing demand gap of long-tail groups.

Featured with informatization, intelligence, and networking, Internet information technology has greatly facilitated the communication of information [2]. The Internet era has greatly weakened the limitations of time and space on people, turned the world into a global village where individuals can communicate with each other even if they are thousands of miles apart. Indeed, the Internet has greatly improved the ability of mankind to understand and change the world, created new platforms for public services and new ways of mass communication. Xu, Ju [2] also introduced the three characteristics of Internet communication: diffusion, openness, and interaction. (1) Diffusion refers to the subversive changes in the speed and breadth of the spread of information in the Internet age, compared to traditional communication networks. (2) Openness is a natural attribute of the Internet, which is reflected in the dissemination of information, there is no spatial distance, no time difference, and no physical barrier. (3) Interactivity refers to the fact that countless intelligent brains and countless intelligent machines from all over the world are gathered on the same network platform for extensive participation, in-depth communication, mutual learning, and interaction. It can be seen that the development of Internet communication technology and the increase in penetration rate can greatly improve the quality of the information environment of listed companies, and finally reduce the information opacity between corporates and investors. And thus, it is reasonable to employ the Internet penetration rate as a proxy for the corporate information disclosure environment.

The second type of literature: the relationship between information disclosure environment and corporate innovation. A large number of studies have shown that the corporate information disclosure environment is very important to external investors and directly affects the future innovation activities of companies. Bellora, Guenther [7] believe that innovation is one of the firms’ core competitiveness. Different from traditional physical investment, R&D investment is featured with intangibility, firm-level heterogeneity, and income uncertainty [16]. Information disclosure of corporate innovation assets is an important basis for investors to evaluate corporate growth and investment value [17]. External investors, other stakeholders, and the public all have a strong demand for corporate innovation information. The information asymmetry and agency costs are the fundamental reasons that distort managers’ decisions and cause companies to deviate from the optimal level of innovation [5, 18]. Cohen et al. [6] provide evidence that the information disclosed by listed companies has continuous predictability on the R&D expenditure of firms. Investors can use investment portfolio strategies to obtain nearly 11% of annual excess returns. Brynjolfsson [19] shows that Internet plays an essential role in reducing the cost of information search and copying, as well as obtaining information resources in corporate innovation activities. The availability of Internet information is an important factor influencing corporate innovation. For example, Kong et al. [20] use the withdrawal of Google search in the Chinese mainland market as an exogenous shock. They show that the innovation input intensity and output quality of domestic firms, which rely heavily on foreign technology, have suffered a negative shock.

2.2 Research hypothesis

The increase in Internet penetration can improve investors’ information acquisition and supervision capabilities, improve the quality of information disclosure. In this situation, investors’ tolerance for corporate innovation failure has also increased accordingly [21]. Especially in Chinese companies with a high degree of equity concentration, large shareholders usually hold corporate stocks for a long time and have a high tolerance for short-term failures [5, 22]. Therefore, in the context of increasing Internet penetration, the risk of management being dismissed due to innovation failure will be reduced, which provides favorable conditions for improving the management’s subjective willingness to innovate. Besides, for investors, with the alleviation of the feeble quality of information, the uncertainty of corporate earnings will also decrease [23], and correspondingly, corporate financing costs will also decrease. Then, it will help ameliorate corporate financing constraints [24], provide financial support for corporate innovation [16], and improve corporate innovation enthusiasm. Formally, the first hypothesis is stated as follows:

  1. H1a: The increase in Internet penetration rate will significantly increase the level of innovation investment of firms.

Generally, the improvement of the information disclosure environment is a double-edged sword for companies. In addition to the above benefits, it will also bring corresponding proprietary costs (costs caused by competitive sensitive information disclosure) [7]. Economic theory shows that proprietary costs are an important deterrent factor for a company’s full voluntary disclosure [25]. The increase in Internet penetration helps improve the corporate information disclosure environment and reduce the asymmetries between the company and the capital market. In this process, the capital market is likely to obtain important information such as expected returns from corporate innovation activities in the future, causing the company to face information leakage risk to competitors. Jia [26] argues that when competitors use this competitive sensitive information to make entry or exit decisions, it will greatly weaken that target company’s competitive advantage and lead to unnecessary competition or imitation. Therefore, the increase in Internet penetration is likely to make companies be more inclined to obtain information from competitors through the Internet and become imitators or low-cost manufacturers of innovative products. This will discourage their initiative to innovate. Brown, Martinsson [16] also believe that a good information disclosure environment makes it more difficult for companies to conceal their innovative activities from the market. Once the important information related to the research and development is leaked to competitors, it will incur serious negative costs. When these negative costs are large enough, corporate transparency will seriously hinder corporate innovation. The above theory and logical reasoning are captured in the following opposite hypothesis:

  1. H1b: The increase in Internet penetration rate will significantly reduce the level of innovation input by firms.

3. Materials and methods

3.1 Data

The sample selection procedure is as follows: (1) Excluding financial listed companies; (2) Deleting firms with missing information; (3) To eliminate the influence of outliers, the continuous variables are winsorized at the 1% and 99% levels. After the above sample screening, the final sample has 27,234 firm-year observations from 2003 to 2017.

The city-level Internet penetration rate data and some particular firm-level indicators come from the Chinese Research Data Services (CNRDS) Platform database. R&D data comes from the WIND database, and other financial data mainly come from the China Securities Markets and Accounting Research (CSMAR) database. For specific definitions of variables, see S1 Appendix.

3.2 Major variable construction

3.2.1 Explained variable.

The variable of interest is the firm’s R&D investment intensity (R&D). This paper focuses on the impact of the information disclosure environment on corporate innovation motivation. Therefore, it is not appropriate to use innovation output as an explained variable. By following the common practices of the existing literature and the intensity of innovation investment R&D is used to proxy innovation motivation [27–29]. R&D is specifically defined as the ratio of the company’s annual R&D expenditure to the total assets at the beginning of the year [27, 30–32].

3.2.2 Explanatory variables.

The explanatory variable Internet is the Internet penetration rate of the city where the listed company is located. Earlier research provides unified proxies for the corporate information disclosure environment. This is the first paper to employ Internet penetration rate to proxy the corporate information and attempt to explore the impact of regional Internet penetration rate on corporate innovation behavior. While Shen, Yuan [4] have found that the Internet penetration rate in the area where the company is located will promote corporate innovation, in their research, the Internet penetration rate is only used as a control variable, not a core explanatory variable. More importantly, due to the unavailability of data, they only collect the Internet penetration rate at the provincial level. This indicator is relatively crude, and difficult to capture the variation in Internet penetration rates among listed companies. Our paper is the first one to use a city-level Internet penetration rate. Our Internet penetration indicator can be more precise to examine the impact of the Internet penetration rate on corporate innovation.

3.2.3 Control variables.

Following previous literature [4, 33, 34], the control variables include firm size (Size), firm age (Age), capital intensity (Capital), Tobin’s Q (Q), return on assets (ROA), financial leverage (Lev), free cash flow (CF), firm growth (Growth), financial distress (Loss), CEO duality (Dual), board size (Board), the proportion of independent director (Indd), major shareholder’s shareholding ratio (TOPHOLD) and the Herfindahl-Hirschman Index (HHI). S1 Appendix presents detailed definitions of all variables.

3.3 Multiple regression model

The following multiple regression model is used to examine the impact of Internet penetration rate on corporate innovation: (1) (2) where the explained variable is corporate R&D investment intensity R&D, following Koh, Reeb [29], the missing values of R&D indicators is replaced with 0. The explanatory variable is the Internet penetration rate Internet. Controls represent control variables. Since R&D investment intensity (R&D) is a non-negative, continuous value censored variable (Censored variable), there is a zero-point censoring problem, so it is more appropriate to use the Tobit model for regression analysis [35, 36]. α represents three types of fixed effects, including the year(ηt), industry(ηind) and region(ηarea) level fixed effects respectively [30, 37, 38]. The fixed effects at different levels can effectively alleviate the endogenous problems caused by missing variables [37]. Besides, to ensure the robustness of the conclusions, our paper follows Petersen [39], and cluster the standard errors by the firm [40–42]. If H1a (H1b) is valid, then β1 is positive (negative).

4. Results

4.1 Summary statistics

The summary statistics of the sample are presented in Table 1. The mean of R&D is 0.010, and the standard deviation is 0.019, indicating that there is considerable variation of R&D among firms. This is consistent with the findings of Lu, Dang [43], Ni, Zhu [30]. The mean (median) of the Internet is 0.377 (0.282), and the standard deviation is 0.412, suggesting that the Internet penetration rate of prefecture-level cities in China is about 37.7%, and there is a large variation of Internet penetration rate in different regions, which helps to identify the impact of Internet penetration rates of different regions on corporate innovation. The summary statistics of control variables are consistent with previous literature [31].

thumbnail
Download:
Table 1. Summary statistics.

https://doi.org/10.1371/journal.pone.0247549.t001

4.2 Baseline results

Table 2 presents the baseline results. In addition to the full Eq (2), a simplified regression model without control variables is designed as Eq (1). Consistently across the two columns, the coefficients of the Internet are positive and significant at the 1% level. The results are economically significant. Using Eq (2), the coefficient of the Internet is 0.001, suggesting that for every additional unit of the Internet, a firm’s R&D is, on average, 0.001 bigger than before. With the mean of R&D at 0. 010, a 0.001 increase represents approximately a 10% increase in R&D relative to a mean firm. The results support H1a, which hints that with the improvement of the Internet environment in the city where the firm is located, external investors may obtain incremental information, the information dissymmetry and financing constraints then will be alleviated, and even external investors are likely to increase their tolerance of management innovation failure. These factors finally stimulate firms’ motivation to invest in innovation.

thumbnail
Download:
Table 2. The impact of Internet penetration rate on corporate innovation.

https://doi.org/10.1371/journal.pone.0247549.t002

In terms of control variables, Size, ROA, CF, Board, Indd and HHI are positive and significant at the 1% level, suggesting that R&D investment intensity is greater if a firm has a larger size, higher profitability, richer free cash flow, larger board size, tighter independent director supervision, and face more fierce market competition. In contrast, the coefficients of Age, Capital, Q, Lev, Growth, Loss, Dual, and TOPHOLD are negative and significant at the 1% level, indicating that firms with a higher degree of these factors, result in a lower intensity of R&D investment. The empirical results of these control variables are consistent with previous literature [35, 36, 40].

4.3 Robustness checks

4.3.1 Alternative measures of key variables.

For robustness, firstly, the explanatory variable (R&D) is redefined as R&D/Sale. R&D/Sale represents the ratio of expenditure R&D expenditure to annual sales revenue [35, 37, 41, 43]. Next, the explanatory variable (Internet) is re-measured as Internet2, which represents the number of Internet broadband access users in prefecture-level cities divided by the annual average population in each city. The robustness test results of this part are shown in columns (1) and (2) in Table 3. The results in columns (1) and (2) in Table 3 show that the coefficients of the Internet are positive and significant at the 1% level. Overall, using alternative core variables does not change our baseline findings in Table 2.

thumbnail
Download:
Table 3. Robustness checks: Alternative measures of key variables and empirical specifications.

https://doi.org/10.1371/journal.pone.0247549.t003

4.3.2 Alternative empirical specifications.

The baseline results in Table 2 employ the Tobit model for regression analysis. For robustness, the full Eq (1) is re-run by using the ordinary least square (OLS) regression. Besides, the baseline results from control industry, city, and year fixed effects and cluster standard errors at the firm level. And then, these fixed effects are changed into city × year and industry ×year, which can control the unobservable heterogeneities over time [38, 44]. For example, the economic development level, intellectual property protection, and the implementation of innovation policies in different regions, etc. There was one caveat: after controlling the city×year fixed effects, STATA cannot be run and report the error that “mat size too small”, so city×year is replaced with province×year fixed effects. Finally, to avoid the clustering problem caused by pool data [36], the standard deviations clustering method is further changed into firm×year level. The specific results of this part are shown in columns (3)-(5) in Table 3 respectively. The results in columns (3)-(5) in Table 3 show that the coefficients of the Internet are consistently positive and significant at the 1% level. Hence, the findings are robust to alternative empirical specifications.

4.3.3 Propensity Score Matching (PSM).

The Internet penetration indicator is relatively exogenous to firms, as neither an individual firm can determine regional technology development. Therefore, the endogeneity between the Internet and R&D should be minimal. To further alleviate omitting variables concern, the propensity score matching method (PSM) test is conducted.

This study follows Ni, Zhu [30] and He et al. [40] to divide all samples into two categories: treatment group and control group. Specifically, the median of the Internet penetration rate (Internet) is used as a benchmark and classify the full sample into high vs. low subsample by using this benchmark. Treat equals to 1 if Internet value is greater than the benchmark, which means treatment group, and 0 otherwise.

Then, our paper chooses the nearest neighbor matching method of 1:1, keep the set of control variables in Eq (1). and use a logistic regression of PSM’s first step to calculating the propensity score. In the process, It is required that the distributions of treatment and control firms meet the common support assumption [45]. Specifically, the propensity score of any observation in the common support set is located in the intersection of the treatment group and the control group. The matching method with replacement is adopted to find the control group sample with the closest propensity score for each firm in the treatment group [40, 46, 47]. The PSM procedure mitigates the impact of unobservable variables on our findings.

Fig 1 illustrates the distribution of propensity scores before and after matching. (a) describes the distribution of propensity scores before matching; (b) describes the distribution of propensity scores after matching. The horizontal axis is the propensity score and the vertical axis is the kernel density. By comparing the distribution of propensity scores before and after matching, it can be found intuitively that the propensity score matching (PSM) method significantly corrected the propensity score deviation between the treatment group and the control group. The matching result is more in line with expectations.

thumbnail
Download:
Fig 1. Distribution of propensity scores before and after matching.

https://doi.org/10.1371/journal.pone.0247549.g001

Fig 2 reports the comparison of the difference of covariates before and after matching. Except for Age, the deviations of other variables after matching are less than 5%, indicating that the balancing assumption is met.

thumbnail
Download:
Fig 2. The comparison of the difference of covariates before and after matching.

https://doi.org/10.1371/journal.pone.0247549.g002

The results of PSM are presented in column (1) in Table 4. The coefficient of the Internet is positive and significant at the 1% level, which is similar to the results in Table 2. After alleviating the endogenous problem, our result still robust.

thumbnail
Download:
Table 4. PSM and Heckman.

https://doi.org/10.1371/journal.pone.0247549.t004

4.3.4 Heckman models.

The problem of missing or zero value of R&D expenditure of some firms will cause sample selection bias. To weaken the impact of this endogenous problem on the research conclusions, the Heckman two-stage method is used to modify Eq (1) by following Shi, Wang [48]. A probit regression is firstly run by using Eq (3) to get the inverse mills ratio IMR. R&D dummyi,t+1 is a dummy variable, which equals 1 if R&D investment is greater than 0, and 0 otherwise. Controlsi,t are consistent with Eq (1). In the second stage of Heckman models, the selected sample of R&D dummyi, t+1 equaling to 1 was used to perform the ordinary least square (OLS) regression on Eq (4), and other variables were still consistent with Eq (1). The empirical result of Heckman’s two-stage method is shown in column (2) in Table 4.

(3)(4)

The result of the Heckman two-stage model is reported in column (2) in Table 4. The coefficient of the inverse Mills ratio IMR is significantly positive, indicating that the R&D intensity data does exist as a sample selection problem. The Heckman two-stage method mitigates the related endogeneity. The coefficient of the Internet is 0.005 (t = 4.67), which is still significantly positive at the 1% level, and consistent with the baseline result. It is confirmed once again that the increase in the penetration rate of the regional Internet helps to promote the R&D investment intensity and stimulate corporate innovation motivation.

4.4 Potential mechanisms

The information disclosure environment is an important factor that affects corporate innovation. For external investors, the perception of the corporate information disclosure quantity and quality will directly affect their recognition of innovative projects, and the firm’s financing behavior. Moreover, financing constraints and investors’ tolerance for management innovation failure will also be shaped. Specifically, with the increase in Internet penetration, investors’ information acquisition and supervision capabilities will be improved, the informational asymmetries between investors and management will be further alleviated. For one thing, with the reduction of corporate unsymmetrical information, the risk of corporate expected returns is reducing [23], which helps to alleviate corporate financing constraints [24] and provides financial support for corporate innovation [16]. On the other hand, investors’ tolerance for corporate innovation failure will also increase accordingly [21]. So, two possible underlying mechanisms are considered for the impact of Internet penetration rate on corporate innovation: "financing constraints" and "tolerance of innovation failure".

4.4.1 Financing constraints.

The first key feature of innovation activities is the high risk of failure and uncertainty of output [31, 49]. This feature of the innovation is likely to induce information asymmetry, potential moral hazards, and then lead to serious financing constraints for innovation activities [50]. Long-term nature is another important feature of innovation. Innovation is a typical long-term high-intensity capital-intensive investment that requires a large amount of capital investment. Schumpeter [51] points out that as a resource-consuming activity, innovation is an organic combination of creativity and capital. Without the support of capital, creativity often cannot form effective production. Subsequent literature also generally confirms that the financing constraints faced by firms will restrict their innovation activities [35, 50, 52]. When internal funds are insufficient to support innovation projects and external financing is severely restricted, corporates have to postpone or abandon innovation. Therefore, financing constraints directly limit the initiative and intensity of innovation and become the primary problem hindering corporate innovation [53, 54].

However, the increase in Internet penetration rate can help investors more conveniently obtain corporate information, evaluate corporate innovation investment projects, and finally decide whether to invest or not. For external investors, especially creditors, the mitigation effect of Internet penetration rate on the degree of asymmetric information is undoubtedly important protection. Jiang et al. [55] argue that the improvement of creditor protection has a positive effect on corporate innovation; and the increase in long-term borrowing and commercial credit is an important way to promote corporate innovation. Especially in an imperfect capital market, the increase in Internet penetration is better to help companies obtain effective funding to support innovation from external channels. The "financing constraint" mechanism is examined by empirically testing the impact of Internet penetration rate on "credit rating (CR)" and "financing constraint (SA index)".

The common corporate financing constraints indicators include the KZ index [56], the WW index [57], and the SA index [58]. However, the KZ index and the WW index contain many endogenous financial variables, such as cash flow, financial leverage, etc. [58–60]. The SA index can weaken this endogenous problem to a certain extent. Hadlock, Pierce [58] construct the SA index based on firm size and age, which has a small variation over time and are relatively exogenous. This index has the advantage of describing the characteristics of financing constraints from a long-term perspective. Besides, existing literature suggests that the SA index can reasonably measure the degree of financing constraints of Chinese firms [50, 61]. Therefore, the SA index is chosen to proxy corporate financing constraints. (5) Where Total Assets = log (total assets (unit: 1 million yuan)), Age = the number of years the firm has been on the list. The larger the SA, the higher the level of financing constraints of the firm.

Additionally, it can be seen that a more direct evidence of the reduction of financing constraints is the reduction of corporate debt financing costs. Therefore, a new empirical test is added by following Ferrer et al. [62]’ s debt financing costs indicator, Debtcost, which is computed as the ratio of financial expenses to the average corporate debt in year t and year t-1.

The results are presented in Table 5. in columns (1) and (3), the coefficients of Internet are 0.068 (t = 2.29) in columns (1) and -0.002 (t = -2.33) in columns (2), both significant at the 5% level. Additionally, the coefficients of Internet in columns (3) is -0.006 (t = -4.40), negative and significant at the 1% level. The above results indicate that the "financing constraint" mechanism is supported. Specifically, the Internet penetration rate not only improves firms’ credit rating, reduces corporate debt financing costs, but also alleviates the financing constraint directly. It can be seen that the improvement of the information disclosure environment has significantly reduced the level of corporate financing constraints, and provided financial support for corporate innovation investment.

thumbnail
Download:
Table 5. Transmission mechanism.

https://doi.org/10.1371/journal.pone.0247549.t005

4.4.2 Tolerance of innovation failure.

In general, firms are often reluctant to invest in innovation activities, because these projects tend to have a high risk of failure, and the innovation process will also be interfered with by some unpredictable and heterogeneous events [62]. And thus economic theory generally believes that tolerance to failure is a crucial factor in stimulating innovation [63]. Innovation activities need a higher degree of tolerance for failure [49]. External investors’ tolerance for innovation failure is an important factor that affects corporate innovation. For example, Tian, Wang [63] show that IPO firms supported by venture capital (VC) with high-risk tolerance have a higher level of innovation. Also, investors’ tolerance for innovation failure is not static. Financing constraints and career considerations are important factors in reducing the tolerance of venture capital firms (VC) to corporate innovation failure. Information disclosure environment is another factor that impacts the tolerance of investors to corporate innovation. Investors’ perception of the quality of corporate information disclosure will directly affect their recognition of corporate investment behavior and innovation projects, finally will shape their tolerance for management innovation failure. Therefore, the Internet penetration rate (as the proxy of information disclosure environment) is likely to have an important impact on corporate innovation through "innovation failure tolerance" mechanism.

This study follows Aghion et al. [64], Luong et al. [65], and Zhu et al. [42] to use the mandatory change of managers-performance sensitivity to measure external investors’ tolerance to innovation failure. Mandatory change events include "dismissal", "resignation", and "personal reasons" etc.; Non-mandatory change events include "retirement" and "term expiration", "controlling change", "health reasons", "improving the corporate governance structure", "involved in the case", and "end of agency". The coefficient of the interaction term (Internet×Droa) is expected to be significant and positive, which means that the Internet penetration rate has weakened the sensitivity between manager’s forced change and performance. External investors have increased their tolerance for innovation failure. On the contrary, they have reduced the tolerance for innovation failure.

In addition, if investors’ tolerance for management’s innovation failure increases, then the level of management’s risk taking will rise accordingly. So two common management’s risk taking indicators are used to test the "tolerance of innovation failure" channels. Bernile et al. [66] argue that the ex post realized volatility of equity returns should also depend on CEOs’ attitude toward risk. They use Stock Volatility to proxy management risk taking. Stock Volatility is annualized volatility of daily stock returns over the fiscal year [66].Yu et al. [67] use the volatility of corporate earnings to measure the level of risk taking, σ(ROA), specifically: (6) (7)

Where, i represents the corporate, and n is the year in the observation period, with a value of 1–3. Xn is the total number of corporate in a certain industry, and k is the kth corporate in the industry. When calculating the volatility of ROA to measure the level of corporate risk taking, the calculation is based on the rolling year. Since the term of top management of listed companies in Chinese country is generally 3 years, this paper uses 3 years (N = 3) as an observation period for rolling calculations [67].

The results are presented in Table 5. in columns (4)—(6) in Table 5. The coefficient of the interaction term (Internet×Droa) in columns (4) is 0.004 (t = 1.77), which is significantly positive at the 10% level. It shows that the increase in Internet penetration weakens the manager’s forced change-performance sensitivity, and improves investors’ tolerance for innovation failure. The coefficients of Internet is 0.024 (t = 2.32) in columns (5), positive and significant at the 5% level. The coefficients of Internet is 0.004 (t = 2.68) in columns (6), positive and significant at the 1% level. It shows that the increase in Internet penetration improves managerial risk-taking. The above results support the "innovation failure tolerance" mechanism.

4.5 Additional analysis

4.5.1 The effect of Internet penetration on R&D efficiency and total factor productivity.

Corporate innovation is a long-term and complex process, and R&D investment is only part of the whole process. Not until transformed into actual achievement, will R&D investment create value for corporate and the national economy. Therefore, our paper further examines the impact of Internet penetration on R&D efficiency (Inno_Eff), namely, the output efficiency of R&D investment. R&D efficiency(Inno_Eff) is used as an indicator to measure corporate innovation capacity by following Zhao et al. [68], Pan et al. [38], and Quan, Yin [66]. It represents the number of patents brought by unit R&D investment. Besides, taking the lag of patent applications into account, the R&D efficiency is measured as the ratio of the number of patent applications in year t+1 to the R&D investment in year t. In particular, for firms whose R&D expenses and the number of patent applications are both 0, Inno_Eff equals 0; for firms whose R&D expenses are 0 but the number of patent applications is not 0, the value of this variable is recorded as missing [38]. The specific definition of Inno_Eff is in S1 Appendix.

Our paper also concerns whether the Internet penetration will promote corporates’ total factor productivity (TFP) level by increasing the R&D intensity. In the context of Chinese research, the OP method has more advantages than the LP method and the ACF method. Therefore, the OP method is used to estimate the total factor productivity (TFP) [67] and use TFP as the explained variable to examine the impact of Internet penetration on firms’ total factor productivity (TFP).

The results are presented in Table 6. in columns (1) and (2), the coefficients of the Internet are 0.004 (t = 2.59) and 0.090 (t = 4.35) respectively, which are both significantly positive at the 1% level. The above results show that the increase in Internet penetration rate not only increases the R&D intensity, but also further improves the innovation output efficiency and production efficiency of firms, and creates a good information technology environment for firms’ long-term development.

thumbnail
Download:
Table 6. Internet penetration, R&D efficiency, and total factor productivity.

https://doi.org/10.1371/journal.pone.0247549.t006

4.5.2 The effect of Internet penetration on human capital.

In corporate innovation investment expenditure, usually, more than 50% is used to pay R&D personnel’s salary [69]. Their R&D activities can create intangible assets or “new knowledge” with more future benefits for the firm. Since this knowledge is contained in the human capital of R&D personnel and cannot be measured, once the R&D personnel is lost, this knowledge will leave the firm and cause irreversible losses to it. Therefore, innovation activities have high adjustment costs [50]. As the direct subject of the implementation of corporate innovation activities, the proportion of corporate R&D personnel and their treatment level is crucial to corporate innovation. With the continuous gathering of senior R&D personnel on the firm level and their continuous improvement of salaries, it is bound to promote the recombination of high-end elements necessary for innovation and the continuous improvement of innovation capabilities. This study further empirically tests the impact of Internet penetration on the proportion of corporate R&D personnel and their salary, to verify whether the information disclosure environment has improved corporate innovation capabilities through the recombination of high-end elements. Since it is difficult to obtain salary data for R&D personnel, by following Zhu et al. [70], the logarithm of the firm’s per capita salary is used to measure whether a firm uses human capital with higher levels of knowledge and skills. The specific results are shown in columns (1) and (2) in Table 7.

thumbnail
Download:
Table 7. Internet penetration and human capital.

https://doi.org/10.1371/journal.pone.0247549.t007

The results are reported in Table 7. in columns (1) and (2), the coefficients of the Internet are 0.722 (t = 4.31) and 0.227 (t = 12.05) respectively, which are both significantly positive at the 1% level. The above results indicate that the Internet penetration rate has significantly improved the human capital environment of firms, promoted the concentration of high-end R&D elements (R&D personnel) at the firm level, increased the salary incentives for R&D personnel, and created a favorable environment for corporate to actively engage in innovation activities.

5. Conclusions

Innovation is critical for firm value and sustainable economic growth, recent academic research, and public policy discussion have both focused on identifying factors that lead to more and better innovation [62, 71]. Leveraging from the city-level Internet penetration rates data in China from 2003 to 2017, this paper examines the impact of corporate information disclosure environment on innovation investment. Our findings suggest that firms headquartered in cities with high Internet penetration rates are more innovative, i.e. they spend more on R&D. It indicates that the improvement of the information disclosure environment has a positive guiding effect on firms’ innovative behavior. Next, our results remain intact after a battery of robustness checks, such as alternative measures of core variables, alternative empirical specifications, and treatments of endogeneity, both via propensity score matching (PSM) and Heckman two stages model. Then two possible mechanisms, "financing constraints" and "innovation failure tolerance" are further proposed and confirmed. It is found that, with the reduction of information opacity between external investors and firms, the external financing constraints faced by firms have decreased, and investors’ tolerance for management innovation failure has increased significantly, which both stimulate higher innovation motivation of firms. Finally, to shed further light on the economic implications of our findings, two additional tests have been done. It is found that the Internet penetration rate plays an important role in promoting the transformation rate of corporate innovation input and corporate production efficiency. And it also facilitates the concentration of high-end elements required for research and development at the firm level, which means the human capital environment of the firm has been greatly improved.

Given the corroborating results of this paper, our conclusions have several important theoretical and practical significance. First of all, the literature is enriched on the relationship between Internet technology and corporate innovation and the relationship between information disclosure environment and corporate innovation. Different from the previous literature, which mainly starts from the subject of firms, and explores the active Internetization of firms. The Internet penetration rate can help external investors to obtain company information, and promote corporate innovation [4, 20]. Secondly, our paper is the first to use the city-level Internet penetration rate. This indicator can more clearly capture the difference in Internet penetration rates of the listed firm. More importantly, this indicator is relatively exogenous and is less affected by firms’ own investment decisions, thereby helping to alleviate the endogenous problems. Thirdly, in a practical sense, this research responds to the requirements of China’s development by combining two important areas of Internet development and corporate innovation and provides evidence support for measuring the economic value of Internet technology. Our conclusions affirm the positive role of Internet communication technology in guiding the effective allocation of human resources and promoting the healthy and sustainable development of the real economy [10].

The existing literature mainly uses innovation input or innovation output to measure corporate innovation performance [16, 72]. our research only focuses on the impact of Internet penetration rate on innovation input (R&D investment). Therefore, future research can further examine the relationship between Internet penetration and innovation output, such as the number of patent applications and grants. In addition, as global environmental issues become increasingly prominent, green innovation is getting more and more attention, and research on the impact of Internet technology development on environmental innovation is also an important topic worthy of attention.

References

  1. 1. Montoya-Cruz E, Ramos-Requena JP, Trinidad-Segovia JE, Sánchez-Granero MÁ. Exploring Arbitrage Strategies in Corporate Social Responsibility Companies. Sustainability. 2020;12(16):6293.
  2. 2. Xu X, Ju X. Crisis Management in the Internet Age: Evolution Trends, Model Construction and Basic Rules. Management World. 2019;35(12):181–9.
  3. 3. Tan S, Kan S, Cui X. Can Investors’Online Communication with Listed Firms Improve the Information Efficiency of the Stock Market? J Finan Res. 2016:174–88.
  4. 4. Shen G, Yuan Z. The Effect of Enterprise Internetization on the Innovation and Export of Chinese Enterprises. Economic Research Journal. 2020;55(1):33–48.
  5. 5. Chen Q, Ma L, Yi Z. Analyst Coverage and Corporate’s Innovation Performance: The Logic of China. Nankai Business Review. 2017;20(3):15–27.
  6. 6. Cohen L, Diether K, Malloy C. Misvaluing Innovation. Rev Finan Stud. 2013;26(3):635–66. pmid:23719248
  7. 7. Bellora L, Guenther TW. Drivers of innovation capital disclosure in intellectual capital statements: Evidence from Europe. The British Accounting Review. 2013;45(4):255–70.
  8. 8. Li Z. Internet Popularization,Online Government and the Quality of Information Disclosure of “Anti-epidemic” Charity Organizations:Empirical Evidences Based on the Information Disclosure in the Home Isolation Situation. Journal of Central University of Finance & Economics. 2020(5):117–28.
  9. 9. Litan RE, Rivlin AM. Projecting the Economic Impact of the Internet. Amer Econ Rev. 2001;91(2):313–7.
  10. 10. Shi B, Li J. Does the Internet Promote Division of Labor? Evidences from Chinese Manufacturing Enterprises. Management World. 2020;36(4):130–48.
  11. 11. Nie H, Tan S, Wang Y. Innovation, Firm Size and Market Competition: From the Evidence of Firm-level Panel Data in China. The Journal of World Economy. 2008;31(7):57–66.
  12. 12. Zhao Y, Tan Z. E-commerce, Bank Loans and Financing of Small and Medium Enterprises (SMEs). Economic Research Journal. 2012(7):99–112.
  13. 13. Hollenbeck B. Online Reputation Mechanisms and the Decreasing Value of Chain Affiliation. J Marketing Res. 2018;55(5):636–54.
  14. 14. Sun F, Zheng J. Study on the Intelligent Evaluation of Accounting Information Quality of Listed Company Based on the "Internet +". Accounting Research. 2018(3):86–90.
  15. 15. Wang C, Dang C, xu F, Zhong L, Du W. Research on Financial Innovation and Wealth Management in the Context of Internet Finance. Management World. 2018;34(12):168–70.
  16. 16. Brown JR, Martinsson G. Does Transparency Stifle or Facilitate Innovation? Management Science. 2019;65(4):1600–23.
  17. 17. Ma D, Yang Y, Zhang B. Acquiring Land or Making Innovations?——What the Disclosure of Intangible Assets in Chinese Listed Companies Tells Us. China Soft Science. 2012(11):170–82.
  18. 18. Manso G. Motivating Innovation. The Journal of Finance. 2011;66(5):1823–60.
  19. 19. Brynjolfsson E. Wired for Innovation: How Information Technology Is Reshaping the Economy. The MIT Press; 2013. https://doi.org/10.1007/s00253-012-4502-5 pmid:23095941
    • 20. Kong D, Lin C, Wei L, Zhang J. Information Accessibility and Corporate Innovation. Working Paper, Available at SSRN: https://ssrncom/abstract=3291811. 2018.
      • 21. Zhong R. Transparency and firm innovation. J Acc Econ. 2018;66(1):67–93.
      • 22. Alchian AA, Demset H. Production, Information Costs, and Economic Organization. The American Economic Review. 1972;62(2):777–95.
      • 23. Botosan CA. Disclosure level and the cost of equity capital. Acc Rev. 1997;72(3):323–49.
      • 24. Jung SH. The contingent effect of analyst coverage: how does analyst coverage affect innovation and Tobin’s Q? Asia-Pacific Journal of Accounting & Economics. 2016;24(1–2):43–67.
      • 25. Li X. The impacts of product market competition on the quantity and quality of voluntary disclosures. Rev Acc Stud. 2010;15(3):663–711.
      • 26. Jia N. Corporate innovation strategy and disclosure policy. Review of Quantitative Finance & Accounting. 2019;52(1):253–88.
      • 27. Zhao Z, Yang Q, Chen J. Generalist or Specialist: CEO’s Competence Structure and Corporate Innovation. Management World. 2018;34(2):123–43.
      • 28. David P, Hitt MA, Gimeno J. The Influence of Activism by Institutional Investors on R&D. Academy of Management Journal. 2001;44(1):144–57.
      • 29. Koh PS, Reeb DM. Missing R&D. J Acc Econ. 2015;60(1):73–94.
      • 30. Ni X, Zhu Y. Labor Protection, Labor Intensity and Corporate Innovation——Evidence from the Implementation of the Labor Contract Law in 2008. Management World. 2016(7):154–67.
      • 31. Yuan J, Hou Q, Cheng C. The Curse Effect of Corporate Political Resources: Based on the Investigation of Political Connections and Corporate Technological Innovation. Management World. 2015(1):139–55.
      • 32. Li C, Yuan R, Khan MA, Pervaiz K, Sun X. Does the Mixed-Ownership Reform Affect the Innovation Strategy Choices of Chinese State-Owned Enterprises? Sustainability. 2020;12(7):2587.
      • 33. Fang VW, Tian X, Tice S. Does Stock Liquidity Enhance or Impede Firm Innovation? The Journal of Finance. 2014;69(5):2085–125.
      • 34. He J, Tian X. The dark side of analyst coverage: The case of innovation. J Finan Econ. 2013;109(3):856–78.
      • 35. Zhang X, Liu B, Wang T, Li C. Credit Rent-seeking, Financing Constraint and Corporate Innovation. Economic Research Journal. 2017;52(5):161–74.
      • 36. Pan Y, Pan J, Dai Y. Litigation Risk, Judicial Local Protectionism and Innovation. Economic Research Journal. 2015;50(3):131–45.
      • 37. Gu X, Chen Y, Pan S. Economic Policy Uncertainty and Innovation:Evidence from Listed Companies in China. Economic Research Journal. 2018;53(2):109–23.
      • 38. Pan J, Pan Y, Ma Y. Is Cooperation Important? Collaboration Culture and Innovation. J Finan Res. 2019(1):148–67.
      • 39. Petersen MA. Estimating Standard Errors in Finance Panel Data Sets: Comparing Approaches. The Review of Financial Studies. 2009;22(1):435–80.
      • 40. He Y, Yu W, Dai Y, Wang Y. Senior Management’s Professional Experience and Enterprise Innovation. Management World. 2019;35(11):174–92.
      • 41. Hu G, Zhao Y, Hu J. D&O Insurance, Risk Tolerance and Enterprise Independent Innovation. Management World. 2019;35(8):121–35.
      • 42. Zhu B, Zhang X, Zheng X. Multiple major shareholders and corporate innovation. Management World. 2018;34(7):151–65.
      • 43. Lu T, Dang Y. Corporate Governance and Innovation:Differences among Industry Categories. Economic Research Journal. 2014;49(6):115–28.
      • 44. Gormley TA, Matsa DA. Common Errors: How to (and Not to) Control for Unobserved Heterogeneity. Rev Finan Stud. 2014;27(2):617–61.
      • 45. Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70(1):41–55.
      • 46. Bartram SM, Brown G, Stulz RM. Why Are U.S. Stocks More Volatile? J Finance. 2012;67(4):1329–70.
      • 47. Cao X, Wang Y, Zhou S. Anti-corruption campaigns and corporate information release in China. J Corp Finance. 2018;49:186–203.
      • 48. Shi X, Wang A. Impact of the Specialty of Corporate Governance on Innovation: Global Evidence from the Dual-class Structure of Internet Firms. Economic Research Journal. 2017;52(1):149–64.
      • 49. Holmstrom B. Agency costs and Innovation. J Econ Behav Organ. 1989;12(3):305–27.
      • 50. Ju X, Lu D, Yu Y. Financing Constraints, Working Capital Management and the Persistence of Firm Innovation. Economic Research Journal. 2013;48(1):4–16.
      • 51. Schumpeter JA. Theory of Economic Development: An Inquiry into Profits, Capital, Credit, Interest, and the Business Cycle. Harvard University Press, Cambridge1934.
        • 52. Brown JR, Fazzari SM, Petersen BC. Financing innovation and growth: Cash flow, external equity, and the 1990s r&d boom. J Finance. 2009;64(1):151–85.
        • 53. Brown JR, Martinsson G, Petersen BC. Do financing constraints matter for R&D? Europ Econ Rev. 2012;56(8):1512–29.
        • 54. Zhou Y. Research on the Performance, Causes and Governance of Structural Features of Chinese Enterprises’ Innovative Financing Constraints. Management World. 2017(4):184–5.
        • 55. Jiang J, Shen D, Jiang X, Yi Z. Creditor Protection and Corporate Innovation. J Finan Res. 2017(11):128–42.
        • 56. Lamont O, Polk C, Saa-Requejo J. Financial constraints and stock returns. The Review of Financial Studies. 2001;14(2):529–54.
        • 57. Whited TM, Wu G. Financial constraints risk. The Review of Financial Studies. 2006;19(2):531–59.
        • 58. Hadlock CJ, Pierce JR. New Evidence on Measuring Financial Constraints: Moving Beyond the KZ Index. The Review of Financial Studies. 2010;23(5):1909–40.
        • 59. Berkowitz D, Lin C, Ma Y. Do property rights matter? Evidence from a property law enactment. J Finan Econ. 2015;116(3):583–93.
        • 60. Lu S, Chen S. Does Governmental Favoritism Reduce Financing Constraintsof Firms: A Quasi-Natural Experiment from China. Management World. 2017(5):51–65.
        • 61. Dou H, Zhang H, Lu Z. Enterprise Group, Major Shareholder Supervision and Over-investment. Management World. 2014(7):134–43,71.
        • 62. Adhikari BK, Agrawal A. Religion, gambling attitudes and corporate innovation. J Corp Finance. 2016;37:229–48.
        • 63. Tian X, Wang TY. Tolerance for Failure and Corporate Innovation. Rev Finan Stud. 2014;27(1):211–55.
        • 64. Aghion P, Van Reenen J, Zingales L. Innovation and Institutional Ownership. Amer Econ Rev. 2013;103(1):277–304.
        • 65. Luong H, Moshirian F, Nguyen L, Tian X, Zhang B. How Do Foreign Institutional Investors Enhance Firm Innovation? J Finan Quant Anal. 2017;52(4):1449–90.
        • 66. Quan X, Yin H. Chinese Short Selling Mechanism and Corporate Innovation: A Natural Experiment from Chinese Margin Trading Program. Management World. 2017(1):128–44.
        • 67. Olley GS, Pakes A. The Dynamics of Productivity in the Telecommunications Equipment Industry. Econometrica. 1996;64(6):1263–97.
        • 68. Zhao J, Li L, He Y. The Impact of Group Reform on Corporate Innovation. Management World. 2019;35(12):152–63,89.
        • 69. Stigliz JE, Weiss A. Credit Rationing in Markets with Imperfect Information. Amer Econ Rev. 1981;71(3):393–410.
        • 70. Zhu Z, Huang X, Wang H. Does Traffic Infrastructure Promote Innovation?A Quasi-natural Experiment Based on the Expansion of the High-Speed Railway Network in China. urnal of Financial Research. 2019(11):153–69.
        • 71. Romer PM. Endogenous Technological Change. J Polit Economy. 1990;98(5):71–102.
        • 72. Gao H, Zhang J. SOX Section 404 and Corporate Innovation. J Finan Quant Anal. 2019;54(2):759–87.