Research on the Coupling Coordination Degree of Triple Helix of Government Guidance, Industrial Innovation and Scientific Research Systems: Evidence from China

Abstract

The coordinated development among government guidance, industrial innovation and scientific research systems has a profound impact on scientific and technological innovation. By constructing a triple helix evaluation index system covering 3 first-class indicators and 32 second-class indicators, this study calculates the coupling coordination degree (CCD) of the triple helix of government (G), industry (I) and university (U) in China from 2010 to 2020, analyzes its time change trend and spatial regional differences, and discusses macro influencing factors. The findings are as follows: First, the coordinated development of government guidance, industrial innovation and scientific research subsystems is low. Although the CCD of triple helix of GIU has been improved, it is still in a mild imbalance state. The CCD of triple helix of GIU in the eastern region is the highest, and that in the western region is the lowest. The CCD of triple helix of GIU in the central region grows the fastest, and that in the northeast region grows the slowest. Secondly, informationization level, opening to the outside world, urbanization level, market demand and economic development are the main influencing factors of the CCD of triple helix of GIU, but these factors have heterogeneous influences on different regions.

1. Introduction

In recent years, China has promulgated a series of policies to strengthen scientific and technological cooperation among government (G), industry (I) and universities (U), and promoted collaborative innovation among GIU. For example, the “Overall Plan for Systematically Promoting Comprehensive Innovation Reform Experiments in Some Regions” takes “promoting the construction of a technological innovation system with enterprises as the main body and combining government, industry, university and research” as one of its main tasks. “Made in China 2025” puts forward “perfecting the manufacturing innovation system with enterprises as the main body, market as the guide and combining government, industry, university and research”. According to the Global Innovation Index report released by the World Intellectual Property Organization, China’s Global Innovation Index ranking rose from 17th in the world in 2018 to 11th in the world in 2022. In the innovation linkages section, China’s ranking rose from 58th in the world in 2018 to 30th in the world in 2022. Although China’s comprehensive innovation capability has been significantly enhanced in recent years, there is still much room for improvement in collaborative innovation among GIU. Some scholars have analyzed the characteristics of collaborative innovation among GIU in China from the co-patent data, and found that the cooperation between UI is the closest, and it has been strengthening, while the cooperation between GU and GI is weak [1].

China’s regional industrial development and growth poles have their own emphasis. The eastern region is a gathering place of high-tech industries and a highland for the development of intelligent manufacturing in China. According to the China Statistical Yearbook of High Technology Industry, the number of high-tech enterprises in the eastern region is about twice that of the total number of the central region, the western region and the northeast region in 2020. The central region is an important industrial transfer base, with the main task of promoting the industrial transfer from east to west, focusing on engineering machinery, intelligent voice, optoelectronic information, new materials and other fields. The northeastern region is an important old industrial base, dominated by resource-based industries such as steel and petrochemical, but these industries have low added value and large consumption of resources and energy. With the depletion of resources and market changes, the northeastern region is actively carrying out industrial transformation. The industrial base in the western region is weak, with high-energy-consuming industries such as coal, electricity, smelting and chemical industry as the leading industries. Moreover, there are great differences in natural resources and economic development among provinces in the western region, and the provinces with better industrial development have limited spatial radiation effect on neighboring provinces.

In a knowledge-based society, the interaction among GIU provides surging power for innovation-driven development, which produces new institutional and social outputs for the production, transfer and application of knowledge [2]. The government shoulders the macro leading responsibility and encourages the cooperation between universities and industries through policies and measures such as scientific capital investment, patent incentives, innovation subsidies and salary regulation of scientific research personnel [3,4]. Industry shoulders the responsibility of scientific and technological innovation, applying the research and development achievements of universities to the development of new products, and promotes the integration of science and technology with economy. Universities shoulder the responsibility of tackling key problems in scientific research, overcoming common technical problems faced by industries, and provide basic theoretical and technological support for industrial development. The exploratory transformation of universities and their participation in regional cooperation have a positive spillover effect on regional innovation systems [5]. Studies have shown that the trilateral cooperation, network relationship and complementary synergy among GIU can promote innovation and entrepreneurship activities by shaping the environment of innovation and entrepreneurship, thus having a positive impact on regional innovation and entrepreneurship [6]. Therefore, the coordinated development of GIU determines whether the innovation system can operate effectively. Thus, how to judge the degree of coordinated development among GIU is a problem worth studying. Therefore, Leydesdorff put forward the triple helix dynamic indicator [7], which is widely used by other scholars to measure the bilateral relationship between G and I, G and U, I and U and the trilateral relationship among GIU.

Joint research and development of patents and co-authorship of academic papers are important forms of cooperation among GIU and are also the main measures to measure the triple helix of GIU. The effective operation of the triple helix system of GIU depends not only on the intersection of the functions of the three, but also on the exertion of their respective functions. According to the barrel principle, the overall level of a system depends on whether each subsystem in the system develops harmoniously. Therefore, ensuring the synchronization and adaptation of innovation development among GIU is conducive to better cohesion and exertion of the joint efforts of the three as a whole. This study focuses on the perspective of system coordination, by measuring the coupling coordination degree (CCD) among government guidance system, industrial innovation system and scientific research system, analyzes the triple helix relationship among GIU and further discusses the macro influencing factors, making marginal contributions.

The main innovative points of this study are as follows: First, it constructs a triple helix evaluation index system of government guidance, industrial innovation and scientific research, and systematically evaluates the relationship among them. Existing studies have mainly explored the relationship among GIU by extracting relational data from single indicators such as patents and papers. Based on the systematic perspective, this study insists on the unity of whole and part. It not only analyzes the scientific and technological innovation activities of three subsystems—government guidance, industrial innovation and scientific research, selecting individual indicators that can express the scientific and technological innovation activities of each subsystem, but also discusses the overall synergy of the triple helix system by using the CCD model. Secondly, the macroscopic influencing factors of the CCD of triple helix of GIU are found. Existing studies have focused on the status quo of the triple helix relationship among GIU, but the analysis of the reasons for the formation of this relationship is not in-depth, and there is also a lack of discussion on regional heterogeneity. This study not only examines the influence of macro factors such as economic development, informationization level and market demand on the triple helix relationship among GIU as a whole, but also further analyzes the heterogeneous influence of these factors on the eastern, central, western and northeastern regions of China, so as to provide support for putting forward policy suggestions according to local conditions and taking advantage of the situation.

This study is mainly composed of the following parts: The literature review section summarizes the representative studies and their main viewpoints from the aspects of the structure types of the relationship among GIU, the triple helix theory, the measurement of the triple helix of GIU, and the influence of the triple helix on innovation, and puts forward the breakthrough point of this study. In the section on research design, this study first constructs a triple helix evaluation index system of GIU, which includes the first-level indicators of government guidance, industrial innovation and scientific research subsystems. According to the characteristics of scientific and technological innovation activities of each subsystem, 32 second-level indicators are selected. Secondly, this study uses entropy method to calculate the weight of each secondary indicator, and then gets the comprehensive scores of the three subsystems. Finally, this study uses the CCD model to measure the synergy and adaptation degree of innovation and development among GIU. The empirical analysis section demonstrates the overall development trend and regional differences of the CCD of triple helix of GIU in 31 provinces of China, and further tests the macro influencing factors of the CCD of triple helix of GIU. The section of conclusions and suggestions summarizes the main research findings and combines them with the evaluation index system and macro influencing factors, putting forward suggestions to improve the CCD of the triple helix of GIU.

2. Literature Review

According to the degree of government intervention, the structure of GIU relationship can be divided into three types. The first is the “etatistic model”, in which industries and universities are nested in the government, and the government controls the activities of industries and universities. The second is the “laissez-faire model”. The structure of the relationship among GIU shows that the government, industry and university are alienated from each other, there is no intersection among them, and the government does not interfere with the activities of industry and university. The third is the “innovation model”, that is, the “triple helix model”. The structure of the relationship among GIU shows that the functions of government, industry and university overlap with each other, and there is intersection at the interface of the three. Through collaborative innovation, a spiral innovation process is formed [8]. In the triple helix model, each helix of government, industry and universities should not only realize internal development, but also interact with other helices through the exchange of goods and services and complementary functions to achieve common development [9]. New functions such as knowledge production and transfer, market and economy, and regulation and control will be formed at the overlapping boundary among GIU [10]. As the “etatistic model” emphasizes the top-down relationship among GIU and reduces the possibility of bottom-up innovation, this model inhibits innovation. The “laissez-faire model”, which requires laissez-faire policies to reduce government intervention, is considered as shock therapy. Most countries and regions adopt the “triple helix model” and devote themselves to shaping the innovation environment of cooperation among GIU [11]. All countries and regions should adjust the “triple helix model” according to local conditions to enhance the adaptability of this model. In developed countries and regions, the government should formulate public policies to encourage cooperation between universities and industries. In emerging countries and regions, the government should also intervene directly to promote the connection between universities and industries [12].

In the current studies of the triple helix relationship among GIU, the bilateral relationship between industries and universities has attracted more attention. Lopes et al. divided the process of commercialization and industrialization of scientific and technological achievements in universities into seven stages, namely, scientific discovery, invention dissemination by R&D personnel, invention patent application evaluation, patent registration, technology marketing and supply, license negotiation, formal or informal commercialization [13]. Leischnig et al. explored the relationship between alliance management ability and technology transfer in university–industry cooperation by using a fuzzy set qualitative comparative analysis method, and constructed a multivariate configuration to promote technology transfer. The results showed that the alliance management ability of academic institutions had a significant positive impact on the success of technology transfer [14]. Based on the perspective of triple helix theory and a longitudinal case study of a UK–China innovation project, Corsi et al. found that universities play the role of internationalization catalyst in the cooperation between universities and industries, and thought that the third mission of universities is to promote the transformation of academic knowledge into industries [15]. In order to promote the bilateral cooperation between industries and universities, the government should rationally allocate public funds among partners in specific investment forms to establish a cooperative culture [16].

Scholars have made useful attempts towards the collaboration among GIU and the measurement of triple helix. Therein, the most commonly used measure is the triple helix indicator based on information entropy theory. The triple helix indicator can reflect the relationship among GIU, and the calculation of this indicator is mainly based on the subset information of government, industries and universities and the intersection information among different subsets. Therefore, through the triple helix indicator, it can not only present the scientific and technological innovation activities of a single subsystem, but also quantitatively present the bilateral and trilateral cooperation among GIU in scientific and technological innovation [17]. According to the interactive information among GIU, scholars calculated the relative frequency of their overlapping relationship to measure the system integration [18]. Based on the data of Web of Science, Ye et al. [19] analyzed the relationship among GIU and its evolution trend in the United States, Britain, France and other countries from 1971 to 2010 through a symbolic information measure—triple helix indicator. It was found that with the passage of time, the triple helix interaction among the three subsystems of GIU has weakened, but there is heterogeneity among different countries. Similarly, Khan et al. [20] used blogs, news websites and other internet materials as data sources of the relationship among GIU, and comprehensively used network measurement, content analysis and co-word analysis techniques to calculate the triple helix index, and discussed the changing trend of the relationship among GIU in Korea from 1999 to 2009. The results show that there are certain tensions among Korean GIU. For example, the improvement of one bilateral relationship is always accompanied by the deterioration of another bilateral relationship, and vice versa. This situation is most obvious in the bilateral relations between UI and UG. In order to analyze the triple helix relationship among GIU collaborative innovation in China, Zhuang et al. [21] measured the coordination degree of innovation system by measuring the mutual information among GIU in the collaborative innovation system with the help of patent data and a triple helix algorithm. The research shows that the relationship among GIU shows a loose trend. In some regions with strong innovation capacity, the triple helix relationship of collaborative innovation among GIU is not close, while in some regions with weak innovation capacity, the triple helix relationship of collaborative innovation among GIU is close. In addition, some scholars tried to extract the co-occurrence matrix of innovation subjects from co-patent data [22] and co-authored paper data [23] and used social network analysis to explore the triple helix relationship among GIU.

Multi-subject collaborative innovation is conducive to promoting the complementary advantages of each innovation subject, optimizing the allocation of scientific and technological resources and improving innovation performance [24]. The triple helix of GIU is considered as an effective form to promote innovation, and many studies have confirmed that the triple helix of GIU has a significant impact on innovation. First, the triple helix of GIU helps to improve the efficiency of regional innovation. Zhuang et al. [25] measured the regional innovation efficiency of 30 provinces and cities in China from 2012 to 2018 with the help of the DEA method and took it as the dependent variable, and measured the cooperation degree among GIU with the help of triple helix algorithm and took it as the core independent variable, and tested the influence of the triple helix of GIU on regional innovation efficiency with regression model. The results show that the triple helix of GIU has a significant positive impact on the comprehensive efficiency and scale efficiency of regional innovation, but it has no significant impact on pure technical efficiency. Second, the triple helix of GIU contributes to knowledge diffusion. Paswan et al. [26] found that compared with non-cooperative academic papers, academic papers completed by GIU have stronger communication power on social media platforms. Third, the triple helix of GIU helps to support the formulation of innovation policies. Lerman et al. [27] believe that the triple helix of GIU plays an important role in the formulation of three innovation policy standards, namely, the establishment of a cooperation system, the production and transfer of knowledge and the development of urban location factors. Each subject can express its interest demands through formal or informal channels and help policy makers reasonably define policy objectives and standards.

Generally speaking, the existing studies summarize the relationship structure types among GIU, introducing the triple helix theory, discussing the measurement of triple helix of GIU, and testing the influence of the triple helix of GIU on regional innovation efficiency, knowledge diffusion and innovation policy making. Therein, in the measurement of triple helix of GIU, the existing studies generally take patents or academic publications as the data basis, and measure the degree of cooperation among GIU, which is used as the measurement of triple helix of GIU. Although the government, industry and universities have carried out a lot of cooperation in patent research and development and academic research, the synergy among the three is not only reflected in these two aspects, for example, the collaborative innovation among the three in technology transfer, new product research, production and marketing has not been fully reflected. Therefore, using patent data and academic publication data to analyze the triple helix relationship among GIU has some limitations. Different from the existing studies, based on the triple helix theory, according to the functional orientation of GIU in the triple helix theory, the evaluation indicators including financial support, new product research and marketing, patent research and development, academic publishing, technology transfer, science and technology services, etc. are selected to reflect the innovation activities of government, industry and universities more comprehensively. With the help of the CCD model, the triple helix relationship among GIU was comprehensively evaluated. In addition, this study also tries to explore the macro influencing factors of the CCD of triple helix among GIU, and puts forward some suggestions on ways to improve the CCD of triple helix among GIU.

3. Research Design

3.1. Construction of Triple Helix Evaluation Index System

Scientific and technological innovation activities involve basic research, applied research, experimental and development research, industrialization and other links, and each subject has different resources, specialties and functions in different stages of scientific and technological innovation activities. Therefore, scientific and technological innovation are not independently completed by a single subject, but need the cooperation between different subjects. The triple helix theory put forward by Henry Etzkowitz et al. [28] provides a reasonable explanation framework for the cooperation of multiple scientific and technological innovation subjects. According to the triple helix theory, government, industry and university are three important subjects of scientific and technological innovation, which interact and reflect each other constantly, thus contributing to the emergence of new knowledge and technology, and constantly promoting the transformation of scientific and technological achievements from samples to products and from laboratories to markets. On the one hand, the triple helix theory summarizes the independence of government, industry and university in the innovation system and the unique contribution of each subject to the innovation process. On the other hand, the triple helix theory advocates for the organic combination of government, industry and university with different value choices and goal orientation as three interrelated important components in the innovation system, in order to form synergy in the administrative field, production field and knowledge field, showing the development trend of integration of science and technology and economy.

The triple helix structure itself is a system, including government system, industrial system and university system. Therein, the government system produces laws and regulations, infrastructure and public resources, while the industrial system produces goods and profits, and the university system produces knowledge and research [29]. By increasing public financial support and increasing financial expenditure on science and technology, the government has compensated for the defect of broken capital chain in the applied research stage to a certain extent, built a bridge between the basic research stage and the industrialization stage and guided all innovation subjects to smoothly pass through the “valley of death” of scientific and technological achievements transformation. This study selects the fiscal expenditure on science and technology and government funds from the R&D internal expenditure to express the government guidance function. Universities need to undertake R&D projects entrusted and funded by the government and enterprises, and devote themselves to basic research, applied research and experimental development research by virtue of their advantages in gathering innovative resources in laboratory construction, purchase of precision instruments and equipment and introduction and education of scientific research talents, and conduct scientific research around core technical issues to realize knowledge innovation. In this study, innovation input indicators such as the number of full-time research and development personnel in universities, basic research expenditure in universities, and innovation output indicators such as the number of academic papers published in universities and the amount of technology transfer contracts in universities are selected to express the scientific research function. Industry is market-oriented, which not only transforms the research and development achievements introduced from universities into new products that can be mass-produced, but it also obtains income and profits through new product sales. It is also necessary to apply the new technologies and processes introduced from universities to the development and production of new products so as to improve the efficiency of new product development and increase the technical content and added value of new products. In this study, innovation input indicators such as full-time equivalent of R&D personnel of industrial enterprises above designated size, funds for developing new products of industrial enterprises above a designated size and innovation output indicators such as sales revenue of new products of industrial enterprises above designated size and number of invention patent applications of industrial enterprises above the designated size are selected to characterize industrial innovation function.

In order to calculate the weight of each indicator and the comprehensive score of government guidance, industrial innovation and scientific research, firstly, this study normalized each indicator to eliminate the influence of dimensions on subsequent calculation. As the selected indicators are all positive, the positive normalization method is adopted, referring to Formula (1).

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

(1)

In Formula (1), $x_{ij}^{\prime }$ represents the normalized indicator value of item j in row i. $x_{ij}$ represents the original indicator value of item j in row i. $\max (x_j)$ represents the maximum value of item j, and $\min (x_j)$ represents the minimum value of item j.

Secondly, this study uses the entropy method [30] to calculate the weight of each indicator, as shown in Formulas (2)–(4). The first step is to calculate the proportion $p_{ij}$ of the jth indicator in row i, as in Formula (2).

$$p_{ij}=\frac{x_{ij}^{\prime }}{{\displaystyle \sum _{i=1}^m{x}_{ij}^{\prime }}}, i\in [1,m] j\in [1,n]$$

(2)

The second step is to calculate the information entropy value $e_j$ of the jth indicator, as in Formula (3).

$$e_j=-\frac{1}{\ln n}{\displaystyle \sum _{i=1}^m{p}_{ij}\ln p_{ij}}, i\in [1,m] j\in [1,n]$$

(3)

The third step is to calculate the weight ${\omega }_j$ of the jth indicator, as in Formula (4).

$${\omega }_j=\frac{1-e_j}{{\displaystyle \sum _{j=1}^n(1-e_j)}}, j\in [1,n]$$

(4)

Finally, combined with the connotation of triple helix theory, indicator correlation and data availability, this study constructs a triple helix evaluation index system covering 3 first-class indicators and 32 second-class indicators, as shown in

Table 1. Triple helix evaluation index system of government guidance, industrial innovation and scientific research.

First-Class Indicators	Second-Class Indicators	Unit	Weight
Government guidance	Fiscal expenditure on science and technology	100 million yuan	45.67%
	Internal expenditure of R&D funds: government funds	Ten thousand yuan	54.33%
Industrial innovation	FTE of R&D personnel in industrial enterprises above designated size	Person/year	10.01%
	R&D funds of industrial enterprises above designated size	Ten thousand yuan	9.44%
	Number of R&D projects of industrial enterprises above designated size	Item	10.36%
	Number of new product projects of industrial enterprises above designated size	Item	10.64%
	Expenditure for developing new products in industrial enterprises above designated size	Ten thousand yuan	10.15%
	Sales revenue of new products of industrial enterprises above designated size	Ten thousand yuan	10.26%
	Export sales income of new products of industrial enterprises above designated size	Ten thousand yuan	15.26%
	Number of effective invention patents of industrial enterprises above designated size	Piece	12.55%
	Number of invention patent applications of industrial enterprises above designated size	Piece	11.33%
Scientific research	Total full-time research and development personnel in universities	Person/year	2.92%
	Total full-time staff of R&D achievements application and science and technology service in universities	Person/year	6.04%
	Current year’s expenditure of basic research funds in universities	Thousand yuan	5.14%
	Current year’s expenditure of applied research funds in universities	Thousand yuan	4.94%
	Current year’s expenditure of experimental development funds in universities	Thousand yuan	5.33%
	Number of research and development institutions in universities	Unit	3.06%
	Number of basic research projects in universities	Item	3.18%
	Number of applied basic projects in universities	Item	3.21%
	Number of experimental development projects in universities	Item	3.99%
	Number of R&D achievements application projects in universities	Item	5.16%
	Number of scientific and technological service projects in universities	Item	5.41%
	Current year’s expenditure of R&D achievements application funds in universities	Thousand yuan	6.65%
	Current year’s expenditure of science and technology service funds in universities	Thousand yuan	5.74%
	Number of scientific and technological works published in universities	Piece	2.59%
	Total academic papers published by universities	Piece	2.95%
	Academic papers published by universities on foreign academic journals	Piece	4.51%
	Number of invention patent applications in universities	Item	5.09%
	Number of invention patents granted in universities	Item	5.54%
	Total number of technology transfer contracts in universities	Item	5.13%
	Total amount of technology transfer contracts in universities	Thousand yuan	6.71%
	Total actual income of technology transfer in universities in the current year	Thousand yuan	6.71%

. The relevant indicators of government guidance mainly refer to the research of Cheng et al. [31] and Sun et al. [32]. The relevant indicators of industrial innovation mainly refer to the research of Li et al. [33] and Yu et al. [34]. The relevant indicators of scientific research mainly refer to the research of Wang et al. [35] and Zhang et al. [36]. On the basis of existing research, the indicators in this study have been adjusted and expanded. The corresponding data come from National Research Network, China Science and Technology Statistical Yearbook and Compilation of Science and Technology Statistical Data of Universities, and the data time span is from 2010 to 2020.

The comprehensive scores of the first-class indicators of government guidance, industrial innovation and scientific research are the sum of the products of the normalized second-class indicators and the weights of the second-class indicators under the first-class indicators, which are expressed by $G$, $I$ and $U$, respectively.

3.2. Calculation of Coupling Coordination Degree of Triple Helix

The concept of “coupling” originated in the field of physics, which refers to the dynamic process in which two or more systems interact and tend to coordinate [37]. The more connections among different systems, the higher their correlation, and the closer they are to a balanced and coordinated state [38]. At present, the CCD model has been widely used in multidisciplinary fields to reveal the internal relations and dynamic evolution process among systems, such as green urbanization system and green financial system [39], urban economic-social-environmental system [40], and so on.

In order to calculate the CCD of triple helix of government guidance ($G$), industrial innovation ($I$) and scientific research ($U$), it is necessary to calculate the triple helix coupling degree ($C$) and the triple helix comprehensive development level ($T$) respectively. Referring to the experience and practice of Wang et al. [41], this study calculates the CCD of triple helix through Formulas (5)–(7).

$$C=3\times {\left[\frac{G\times I\times U}{{(G+I+U)}^3}\right]}^{\frac{1}{3}}$$

(5)

$$T={\alpha }_{1}G+{\alpha }_{2}I+{\alpha }_{3}U, {\alpha }_1+{\alpha }_2+{\alpha }_3=1$$

(6)

In Formula (6), ${\alpha }_1$, ${\alpha }_2$, ${\alpha }_3$ and are undetermined parameters, which indicate the importance of $G$, $I$ and $U$ to the $T$. According to the triple helix theory, although the innovation functions of government, industry and universities are different, they are all equally important, so the three undetermined coefficients are set as equivalent in the study.

$$CCD=\sqrt{C\times T}$$

(7)

In Formula (7), the value range of CCD is [0, 1]. According to the existing research about different threshold definition standards [42], the CCD can be divided into 10 grades, as shown in

Table 2. Classification standard of CCD.

Range of CCD	Type of CCD
(0.0, 0.1)	Extreme imbalance
[0.1, 0.2)	Severe imbalance
[0.2, 0.3)	Moderate imbalance
[0.3, 0.4)	Mild imbalance
[0.4, 0.5)	Near imbalance
[0.5, 0.6)	Reluctant coordination
[0.6, 0.7)	Primary coordination
[0.7, 0.8)	Intermediate coordination
[0.8, 0.9)	Good coordination
[0.9, 1.0)	Extreme coordination

.

4. Empirical Results

From the time trend, the CCD of triple helix in China increased from 0.276 in 2010 to 0.396 in 2020. Although it showed an overall upward trend, it was still in a mild imbalance state, as shown in Figure 1. Therein, from 2010 to 2013, the average CCD of triple helix in China was 0.281, which represented a moderate imbalance state. From 2014 to 2020, the CCD of triple helix in China improved, with an average increase of 0.347, and the CCD of each year was always in the range [0.3, 0.4), entering a mild imbalance state. Therefore, there are still unbalanced and uncoordinated problems among the GIU systems, which may lead to risks such as poor channels for the transformation of scientific and technological achievements, disconnection between innovation chain and industrial chain, and difficulty in the effective integration of science and technology and economy.

From the corresponding situations of 31 provinces and cities, there are great differences in the CCD of triple helix among different provinces and cities, as shown in Figure 2. The average CCD of triple helix among provinces and cities is 0.323, which is higher than the average level in 14 provinces and cities and lower than the average level in 17 provinces and cities. After calculation, the CCD of triple helix of Jiangsu and Guangdong is close and obviously higher than that of other provinces and cities, with Jiangsu being 0.678 and Guangdong being 0.667, and their GIU systems have reached the primary coordination state. The CCD of triple helix in Beijing, Shanghai and Zhejiang is in the range of [0.5, 0.6), which represents a reluctant coordination state. The GIU systems of the above-mentioned five provinces and cities have reached a coordination state, and different systems are closely related, forming an effective synergy. In contrast, the CCD of triple helix in other provinces and cities is lower than 0.5, which indicates that there are different degrees of imbalance in GIU systems. Therein, the CCD of triple helix in Shandong, Hubei and Sichuan is within [0.4, 0.5), and the system of GIU is on the verge of imbalance. The CCD of triple helix in Anhui, Shaanxi, Liaoning, Henan, Hunan, Tianjin and Fujian is within [0.3, 0.4), and the GIU system is facing a mild imbalance state. The CCD of triple helix in Chongqing, Hebei, Jiangxi, Heilongjiang, Jilin, Guangxi, Shaanxi and Yunnan is within [0.2, 0.3), and the GIU system presents a moderate imbalance state. The CCD of triple helix in Guizhou, Inner Mongolia, Gansu, Xinjiang, Ningxia, Hainan, Qinghai and Tibet is maintained in the range of [0.1, 0.2), and the system of GIU is severely imbalanced.

From the situation of the four regions, the CCD of triple helix in each region is obviously different and presents different development trends, as shown in Figure 3.

The CCD of triple helix in the eastern region is always higher than that in the central region, northeastern region and western region, increasing from 0.383 in 2010 to 0.559 in 2020, with a growth rate of 45.95%. The state of the GIU system gradually improved from mild imbalance to near imbalance, and finally reached a reluctant coordination state. In terms of government guidance, the eastern region has developed economy and strong financial strength, and it can support the development of science and technology through direct and indirect subsidies such as financial allocation, setting up investment guidance funds, issuing scientific and technological innovation vouchers, and increasing the proportion of R&D expenses plus deduction. In terms of industrial innovation, the industrial structure in the eastern region tends to be advanced and rationalized, with high industrial concentration and superior business environment, attracting a large number of high-tech enterprises and researchers, and actively performing innovative activities such as patent research and development, new product research and production and marketing. In terms of scientific research, there are many first-class universities in the eastern region, with abundant scientific and technological innovation talents and perfect scientific research infrastructure. High-quality resources have been gathered in basic research, applied research and experimental development research, and ties with the government and enterprises have been strengthened by setting up national university science parks and collaborative innovation centers. Therefore, the government guidance system, industrial innovation system and scientific research system in the eastern region are relatively developed, and the innovation subjects promote the collaborative innovation of government, industry and university through a reflexive mechanism.

The CCD of triple helix in the central region gradually increased from 0.278 in 2010 to 0.434 in 2020, and the growth rate was the highest among the four regions, reaching 56.12%. Although the state of GIU system has been improved, there is still a risk of imbalance. Specifically, the development of industrial innovation system in central region lags behind the government guidance system and scientific research system. In recent years, the central region has made important contributions in promoting the transfer of industries from east to west, and industrial transfer and technology transfer complement each other. By undertaking some industries in the eastern region, the central region has strengthened scientific and technological cooperation and exchanges with the eastern region, enhanced the ability of technology introduction, digestion, absorption and re-innovation, accelerated the diffusion of technological innovation and the transformation of scientific and technological achievements, and improved industrial efficiency. However, the place where the industry is transferred out usually regards the place where the industry is undertaken as a product-processing base instead of a technology research and development base. In the process of undertaking industrial transfer, the advanced technology resources that the central region can obtain are limited, and the technology input may form a “crowding-out effect” on the technology in this region, and some local technologies are replaced by imported technologies, thus restricting the power of the industry to conduct technical research and new product development to a certain extent.

The CCD of triple helix in northeastern region increased from 0.258 in 2010 to 0.316 in 2020, and gradually changed from moderate imbalance to mild imbalance, with a growth rate of only 22.48%, which is the lowest among the four regions. From 2010 to 2013, the CCD of triple helix in the northeastern region was very close to that in central region, even higher than that in the central region. However, from 2014 to 2020, the gap between the CCD of triple helix in northeastern region and that in central region became bigger and bigger, and their development trends were quite different. The CCD of triple helix in northeastern region remained basically unchanged, while the CCD of triple helix in the central region grew rapidly. In recent years, the northeastern region has been faced with practical problems such as economic downturn, brain drain and relatively lagging business environment construction, which limit the collaborative innovation level of GIU to a certain extent. For example, the economic downturn has caused a shortage of financial resources and relatively insufficient financial investment in science and technology; Brain drain is manifested in the shortage of R&D personnel, low R&D efficiency and difficulty in gathering increments, resulting in insufficient innovation kinetic energy; The construction of business environments is relatively lagging behind, which may hinder the investment promotion of high-tech industries and strategic emerging industries, delay the progress of technology research and development and new product research and production, and restrict the improvement of regional innovation efficiency.

The CCD of triple helix in the western region is always lower than that in the eastern region, central region and northeastern region, increasing from 0.190 in 2010 to 0.261 in 2020, with a growth rate of 37.37%. The state of GIU systems has changed from severe imbalance to moderate imbalance, and there is still a lack of effective synergy among different systems. The main reason for the low CCD of triple helix in western region is that the economic development in the western region lags behind, and that the concentration of high-tech industries and strategic emerging industries is low, and that the resources of scientific and technological innovation in universities are scarce. There are obvious shortcomings in both input and output of scientific and technological innovation, and in both quantity and quality of scientific and technological innovation achievements, which makes the development trend of government guidance system, industrial innovation system and scientific research system lag behind the other three regions.

5. Further Discussion: The Influencing Factors of CCD of Triple Helix

It is found that the CCD of triple helix shows an increasing trend, and there are obvious differences among different regions. What influencing factors have caused this phenomenon and changing trend? In order to answer this question, according to the research results of Li et al. [43], Liu et al. [42] and Cui et al. [44], this study attempts to further explore the influencing factors of the CCD of triple helix from the aspects of economic development, opening to the outside world, infrastructure, market demand, informationization level and urbanization level. In this study, fixed assets investment of scientific research, technical service and geological exploration industry (100 million yuan) is taken as the proxy variable of infrastructure, expressed as Assets. Take the number of internet broadband access users (10,000 households) as the proxy variable of informatization level, expressed as Internet. The total import and export volume (US $10,000) is taken as the proxy variable of opening to the outside world, expressed as Open. The proportion of urban population to the total population (%) is taken as the proxy variable of urbanization level, expressed as Urbanization. The turnover of technology market (10 thousand yuan) is taken as the proxy variable of market demand, expressed as Transaction, and the per capita GDP (yuan) is taken as the proxy variable of economic development, expressed as PERGDP.

The data of each variable come from National Research Network and China Statistical Yearbook, with a time span of 2010–2020, and a total of 341 samples were obtained. The descriptive statistical results are shown in

Table 3. Descriptive statistics.

Variables	N	Min.	Max.	Mean	Std. Error	VIF
Assets	341	0.79	2069.71	175.0258	279.16604	1.731
Internet	341	12.8	3889.99	954.83813	816.201023	2.511
Open	341	31,053.24	109,158,143.7	13,063,189.75	21,561,142.61	2.331
Urbanization	341	0.22215873	0.954744514	0.567970168	0.134664033	2.56
Transaction	341	0.54	63,161,621.97	3,769,112.588	7,805,859.422	1.841
PERGDP	341	13119	164,889.47	53,663.491	27,189.58105	3.834

. The Variance Inflation Factor (VIF) of each variable is less than 10, which shows that there is no serious multicollinearity among each variable, thus ensuring the accuracy of subsequent regression analysis and parameter estimation to a certain extent.

In order to eliminate the influence of different dimensions and enhance the comparability of the results, this study transformed the original data of each variable into the standardized score Z value, and then carried out regression analysis. In this study, a bootstrap method was used to explore the influence of various variables on the CCD of triple helix. Bootstrap sample size was set to 1000, and 95% confidence interval was selected. The regression results are shown in

Table 4. Bootstrap regression results.

Variables	Coefficients	Std. Error	p Value	95% Confidence Interval
				Lower Limit	Upper Limit
Assets	0.012	0.029	0.692	−0.037	0.078
Internet	0.477	0.026	0.001	0.424	0.527
Open	0.282	0.039	0.001	0.217	0.369
Urbanization	0.149	0.022	0.001	0.101	0.188
Transaction	0.243	0.032	0.001	0.202	0.324
PERGDP	0.096	0.035	0.009	0.019	0.16

.

The regression coefficient of Assets to the CCD of triple helix is 0.012, p = 0.692, which is not statistically significant. Infrastructure construction is the carrier of scientific and technological cooperation among GIU. Buying R&D equipment and building production bases are important contents of infrastructure construction. The government encourages industries and universities to improve their infrastructure by implementing policy tools such as adding and deducting R&D expenses and accelerating depreciation. Industry and universities strengthen scientific and technological cooperation by building R&D bases and sharing R&D equipment. However, the fixed assets investment of scientific research, technical services and geological exploration industry in various provinces and cities are generally low, which makes it difficult to form a strong support for the collaborative innovation of GIU. For example, the average fixed assets investment of scientific research, technical services and geological exploration in Jiangsu, Guangdong and Beijing are 68.189 billion yuan, 22.156 billion yuan and 8.77 billion yuan, respectively, accounting for only 1.47%, 0.70% and 1.16% of the total fixed assets investment of the whole society.

The regression coefficient of Internet to the CCD of triple helix is 0.477, p = 0.001, which shows that informationization level has a significant positive impact on the CCD of triple helix. In recent years, the emerging format of “Internet plus” has been maturing day by day. The application of the internet has broken through the barriers of time and space to a certain extent, enhanced the transparency of information, eliminated the problem of information asymmetry and accelerated the efficiency of information transmission among different innovation subjects. For the government, the improvement of informatization level is conducive to improving the allocation efficiency of scientific and technological resources and the accuracy of public scientific and technological service supply. For the industry, the improvement of informatization level is conducive to reducing production and operation costs and improving the efficiency and added value of new product development. For universities, the improvement of the informatization level is conducive to following up the frontier of world science and technology and the major needs of the country, adjusting the research direction of basic research, applied research and experimental development in time, and enhancing the focus of scientific research.

The regression coefficient of Open to the CCD of triple helix is 0.282, p = 0.001, indicating that opening to the outside world has a significant positive effect on the CCD of triple helix. Carrying out import and export trade is an effective method of open innovation. By absorbing and exporting technology, it promotes the global flow of innovation elements and strengthens international scientific and technological cooperation and exchanges. In addition, import and export trade will have certain spillover effects, that is, import and export trade are usually accompanied by transnational transfer of knowledge and technology. Importing regions realize local transformation of foreign technology and scientific and technological achievements through decomposition, imitation and improvement, and integrate them into local industrial chain and innovation chain, so as to enhance the technical content and market value of scientific and technological achievements. In this process, the government needs to create a business environment conducive to open innovation. The industry needs to intensify technology introduction, digestion, absorption and re-innovation, and universities need to evaluate and demonstrate the basic principles of foreign new technologies and new products and their applicability and feasibility in domestic promotion.

The regression coefficient of Urbanization to the CCD of triple helix is 0.149, p = 0.001, which indicates that the improvement of urbanization level can significantly promote the growth of the CCD of triple helix. Urban population agglomeration accelerates the consumption of regional resources, energy and public goods. In order to achieve green, innovative and sustainable development, higher requirements are put forward for technological progress and transformation of scientific and technological achievements. The “triple helix” in the transformation of scientific and technological achievements means that the government, industry and universities have a clear division of labor and cooperate with each other. The government needs to increase financial investment in science and technology to help the transformation of scientific and technological achievements cross the “valley of death”. Universities need to intensify R&D research, devoting themselves to basic research and applied research, and provide guarantee conditions for the source supply of “samples” of scientific and technological achievements. The industry needs to increase investment in new product development funds, promoting scientific and technological achievements from laboratories to markets, and realize the transformation of scientific and technological achievements from “samples” to “products”.

The regression coefficient of Transaction to the CCD of triple helix is 0.243, p = 0.001, which indicates that the CCD of triple helix increases significantly with the increase of market demand. Market demand orientation is a basic principle and value orientation of collaborative innovation among government, industry and universities. With the complexity of technology and the refined development of industrial chain, it is difficult for a single innovation subject to meet the diversified innovation needs. Therefore, the government, industry and universities should cooperate in innovation and actively carry out scientific and technological cooperation to jointly meet the market needs. In addition, differentiated innovation needs are conducive to the formation of new economic growth points and interest relations, and promote the government, industry and universities to strengthen cooperation to achieve common interests. Therefore, the more active the market demand is, the closer the connection among government, industry and universities is.

The regression coefficient of PERGDP to the CCD of triple helix is 0.096, p = 0.009, which shows that the higher the level of economic development, the higher the CCD of triple helix. In terms of government guidance, the better the regional economic development, the stronger the financial strength of local governments, and the more abundant financial funds for supporting regional scientific and technological innovation. In terms of industrial innovation, economically developed areas will have a siphon effect on the innovation resources in surrounding areas, forming resource advantages conducive to the agglomeration of high-tech industries and strategic emerging industries, and promoting the rationalization and upgrading of industrial structure, thus promoting industrial innovation. From the aspect of scientific research, universities mainly participate in solving many theoretical and practical problems in the process of regional economic development by undertaking R&D projects and scientific research projects. Compared with universities in economically underdeveloped areas, universities in economically developed areas can undertake projects with great difficulty, high quotas and large numbers, they have accumulated more experience and technology in the process of completing projects and they have better scientific research strength.

In order to further explore the heterogeneous influence of each variable on the CCD of triple helix in different regions, this study carried out sample regression analysis according to the eastern region, central region, western region and northeastern region, and set the bootstrap sample size as 1000 and selected a 95% confidence interval. The regression results are shown in

Table 5. Regression results of sub-regional samples.

Regions	Variables	Coefficients	Std. Error	p Value	95% Confidence Interval
					Lower Limit	Upper Limit
Eastern region	Assets	0.018	0.029	0.515	−0.03	0.088
	Internet	0.381	0.058	0.001	0.262	0.486
	Open	0.363	0.059	0.001	0.264	0.502
	Urbanization	0.101	0.058	0.093	−0.022	0.214
	Transaction	0.135	0.025	0.001	0.098	0.193
	PERGDP	0.27	0.043	0.001	0.183	0.35
Central region	Assets	0.104	0.099	0.156	0.035	0.412
	Internet	0.119	0.08	0.115	−0.071	0.263
	Open	2.29	0.645	0.001	1.061	3.598
	Urbanization	0.05	0.12	0.676	−0.2	0.266
	Transaction	0.719	0.126	0.001	0.53	1.004
	PERGDP	−0.16	0.194	0.416	−0.554	0.202
Western region	Assets	0.195	0.218	0.333	−0.177	0.704
	Internet	0.271	0.075	0.003	0.124	0.417
	Open	1.87	0.337	0.001	1.262	2.6
	Urbanization	0.118	0.033	0.001	0.051	0.18
	Transaction	0.563	0.142	0.002	0.309	0.852
	PERGDP	−0.135	0.053	0.017	−0.246	−0.035
Northeastern region	Assets	−0.071	0.052	0.123	−0.178	0.04
	Internet	−0.051	0.124	0.671	−0.314	0.171
	Open	1.198	0.13	0.001	0.935	1.46
	Urbanization	0.328	0.112	0.011	0.107	0.558
	Transaction	0.263	0.11	0.02	−0.034	0.42
	PERGDP	0.034	0.043	0.451	−0.062	0.115

.

There are differences in influencing factors of CCD of triple helix in different regions. As far as the eastern region is concerned, the influence of informationization level, opening to the outside world, market demand and economic development on the CCD of triple helix are significantly positive at the level of 1%, while the influence of urbanization level on the CCD of triple helix is significantly positive at the level of 10%, while the influence of infrastructure is not significant. As far as the central region is concerned, the influence of opening to the outside world and market demand on the CCD of triple helix are statistically significant at the level of 1%, while the influence of infrastructure, informationization level, urbanization level and economic development on the CCD of triple helix are not significant. For the western region, informationization level, opening to the outside world, urbanization level and market demand have a significant positive impact on the CCD of triple helix, while economic development has a significant negative impact, while infrastructure has no significant impact. As far as the northeastern region is concerned, the impact of opening to the outside world on the CCD of triple helix is significantly positive at the level of 1%, while the impact of urbanization level and market demand are significantly positive at the level of 5%, whereas the impact of infrastructure, informationization level and economic development are not statistically significant.

6. Conclusions

The triple helix theory is based on a systematic viewpoint, which insists on the unity of whole and part. It not only clarifies the respective contributions of government system, industrial system and university system to scientific and technological innovation, but also innovatively puts forward that the three subsystems should be organically combined to form a joint force to promote scientific and technological innovation. In fact, the size of the joint force depends to a great extent on whether each subsystem develops harmoniously and synchronously. In order to explore the coordinated development of government guidance, industrial innovation and scientific research subsystems, based on the triple helix theoretical analysis framework, this study calculates the CCD among government guidance system, industrial innovation system and scientific research system in 31 provinces and cities in China from 2010 to 2020, analyzing the collaborative innovation relationship among GIU, and further discusses the macro influencing factors of the CCD of triple helix. The following findings were obtained.

First, the coordinated development of government guidance, industrial innovation and scientific research subsystems is low, so that the combination among subsystems is not close enough. Although the overall CCD of triple helix of GIU in China shows a steady growth trend, it is still in a mild imbalance state. There are obvious differences in the development trend of the CCD of triple helix of GIU among different regions. In terms of absolute quantity, the eastern region has the highest CCD of triple helix of GIU and the coordinated development of each subsystem is the best, while the western region has the lowest CCD of triple helix of GIU and the subsystems are still moderately imbalanced. In terms of growth rate, the CCD of triple helix of GIU in the central region has the largest increase, and the coordinated development of each subsystem has improved rapidly, while that in the northeastern region has the smallest increase, and the problems of unsynchronized and uncoordinated development of each subsystem have not been effectively solved.

Second, there are some differences in the influencing factors of the CCD of triple helix of GIU in different regions. For the whole country and the eastern region, the level of informationization, opening to the outside world, urbanization, market demand and economic development have a significant positive impact on the CCD of triple helix. As far as the central region is concerned, opening to the outside world and market demand have significantly promoted the growth of the CCD of triple helix. For the western region, informationization level, opening to the outside world, urbanization level and market demand are the key influencing factors. For the northeastern region, opening to the outside world, urbanization level and market demand significantly promote the growth of the CCD of triple helix.

7. Policy Implications

Combined with the triple helix evaluation index system of government guidance, industrial innovation and scientific research, and the influencing factors of the CCD of triple helix, this study puts forward the following suggestions in order to enhance the CCD of triple helix and promote the collaborative innovation of government, industry and university.

First, strengthen government guidance. Increase financial investment in science and technology, and give full play to the guiding, supplementing and covering role of financial funds in science and technology. Expand the scale of science and technology innovation investment guidance funds, enriching the types of guidance funds, and drive social capital to gather in the field of science and technology innovation. For investment activities supporting scientific and technological innovation, the science and technology innovation investment guidance funds can be used to follow up the investment according to a certain proportion. Set up science and technology innovation risk compensation funds, and build trial and error, fault tolerance and error correction mechanisms. For capital losses and loan supply cuts caused by R&D interruption and failure of scientific and technological achievements transformation, the science and technology innovation risk compensation funds can provide all-round guarantee according to a certain proportion.

Second, promote industrial innovation. Increase investment in new product development and develop new products such as high-end integrated circuits, big data systems and industrial robots. Expand the application scenarios of new products, and promote consumption upgrading with product upgrading. For example, the big data system will be introduced into government departments, and government service platforms such as smart medical care, smart transportation and smart education will be built, so as to improve the public’s “user portrait”, promoting the “number-based governance” of government departments, and realizing scientific public decision making and accurate public service supply. As another example, high-end integrated circuits are applied to smart homes, communication equipment, vehicle-mounted central control SOC, industrial control systems and other fields to promote the iterative development of high-end integrated circuits in practical applications. Encourage related industries to introduce new products and promote the simultaneous upgrading of the industrial chain.

Third, concentrate on scientific research. Enhance the investment in R&D projects. Purchase and update R&D equipment to improve the technical content of scientific and technological achievements and their transformation efficiency. Attract high-level talents to participate in R&D projects by means of agreed annual salary system and project salary system, and increase the proportion of scientific research project personnel funds. Improve the classification and evaluation system of scientific and technological talents in universities. For researchers engaged in basic research, the academic representative system should be implemented, with high-quality research results as evaluation indicators. For researchers engaged in applied research and experimental development research, the number of technology contracts signed and the transaction amount, technology license and transfer, technical services, invention patent authorization, etc. should be taken as evaluation indicators. Encourage universities to carry out scientific and technological cooperation, forming university alliances, building a scientific research cooperation network with resource sharing, complementary advantages, joint training and collaborative innovation, and jointly apply for and undertake major R&D projects.

Fourth, carry out the pilot work of collaborative innovation policy among government, industry and university. Economic development, urbanization level, market demand, degree of opening to the outside world, informationization level, etc., should be taken as the selection criteria for the pilot policy of collaborative innovation among government, industry and university, and the experience of the pilot policy should be summarized regularly and replicated and promoted in a wider scope, so as to give full play to the spatial radiation effect of the pilot policy. The policy pilot should fully cover the eastern, central, western and northeastern regions, and fully consider the regional heterogeneity of selection criteria and influencing factors. For example, provinces and cities with good economic foundation, high urbanization level, active technology market, high degree of opening to the outside world and rapid information transformation should be selected as policy pilots in the eastern region. Select provinces and cities with a high degree of opening to the outside world and active technology market from the central region as policy pilots. Provinces and cities with rapid information transformation, high degree of opening to the outside world, high level of urbanization and active technology market should be selected as policy pilots in the western region. Provinces and cities with a high degree of opening to the outside world, high level of urbanization and active technology market should be selected as policy pilots in the northeastern region.