Measurement and Spatial Difference Analysis of Innovation-Driven Urban Development Levels in Sichuan Province

Liu Fangbo

Xihua University

Zhu Yanting*

Sichuan Provincial Academy of Natural Resource Sciences

Abstract: Based on the connotation and process of innovation-driven development, we have developed a comprehensive evaluation system containing 20 indicators in five aspects, including innovation factors, innovation subjects, innovation environments, innovation outputs, and development performance, to measure the levels of innovation-driven development in Sichuan province. Selecting 21 cities and prefectures in Sichuan province as research objects, we evaluated and measured the innovation-driven development levels of each city and prefecture using the entropy weight method (EWM). According to the evaluation results, the 21 cities and prefectures were divided into four categories depending on their levels of innovation-driven development: advanced-level, high-level, medium-level, and low-level. The results show that there are obvious spatial differences in terms of innovation-driven development levels among cities and prefectures in Sichuan province. Specifically, Chengdu, Mianyang, Panzhihua, and Deyang cities rank among the top four cities because of their advanced and high levels of innovation-driven development, while other cities and prefectures are at the medium and low levels. We also analyzed the innovation-driven development policies and practices of cities and prefectures in Sichuan province, to provide guidance for implementing innovation-driven development strategies in the cities and prefectures in the future.

Keywords: innovation-driven development, entropy weight method, comprehensive evaluation, spatial differences

Introduction

Michael E. Porter, an American academic known as the father of competitive strategy, divides economic development into factor-driven, investment-driven, innovation-driven, and wealth-driven stages (Porter, 1990). In the innovation-driven stage, innovation is the driving force for economic development. Since the reform and opening-up, China has made fruitful achievements in economic development. China’s economy has shifted from a high-speed growth stage to a highquality development stage (Hou, 2019). Regional and urban developments are transforming from factor-driven and investment-driven development to innovation-driven development, making innovation an important driver in improving the competitiveness of regional and urban industries (Liu, Zhou & Jiang, et al., 2019). As early as 2012, China put forward the innovation-driven development strategy and stressed the critical position of scientific and technological innovation in its economic and social development. In 2016, the State Council officially issued theOutline of the National Strategy of Innovation-Driven Development, making overall plans for the implementation of China’s innovationdriven development strategy. Sichuan province, located in southwest China, is a big province with a large population, vigorous economic development, and rich scientific and educational resources. It plays a crucial role in China’s overall development. With the comprehensive implementation of the innovation-driven development strategy, Sichuan province has seen continuous improvements in independent innovation capability, optimization and upgrading in industrial development, and highquality advancement in economic development. Therefore, innovation as the driving force of the province’s endogenous growth is playing an increasingly important role. At the new stage of historical development, China proposed to “uphold the central role of innovation in its modernization drive and made self-reliance in science and technology the strategic pillar for the country’s development” at the fifth plenary session of the 19th Central Committee of the Communist Party of China (CPC). Sichuan province actively implemented the plans of the CPC Central Committee. In the first year of the 14th Five-Year Plan, Sichuan province issued theDecision of the CPC Sichuan Provincial Committee on Furthering Innovation-Driven High-Quality Development, taking innovation as the driving force for its high-quality development. Against this backdrop, it is of great significance to measure and horizontally compare the innovation-driven development of all cities and prefectures in Sichuan province, to clarify the priorities for implementing innovation-driven development strategies in different areas. Moreover, it is also important to effectively enhance innovation-driven development levels of various regions and build Sichuan into a model province of innovation-driven development, thus providing references for the high-quality development of the whole country.

Literature Review

In terms of the measurement of innovation-driven development levels, the international focus is on the evaluation of innovation capability. F. M. Scherer (1982) and Luis Suarez-Villa (1990) measured innovation capability via patent statistics. Jeffrey L. Furman, Michael E. Porter, and Scott Stern, et al. (1990) established an index system based on the infrastructure and environmental conditions of innovation to evaluate innovation capability. The Global Innovation Index (GII), jointly released by the World Intellectual Property Organization (WIPO), Cornell SC Johnson College of Business, and the Institut Européen d’Administration des Affaires (INSEAD) has made continuous measurements of the innovation capacity of many global economies from the perspectives of innovation input and output since 2007 (Cornell SC Johnson College of Business, INSEDA & WIPO, 2020). The European Innovation Scoreboard (EIS), released by the European Commission, measures innovation performance through comprehensive innovation indicators (World Economic Forum, 2019). The evaluation system includes four first-level indicators (framework conditions, investments, innovation activities, and impacts), 10 secondlevel indicators, and 27 third-level indicators, and mainly carries out the dynamic evaluation of the innovation capacity of EU member states. In addition, the Global Competitiveness Index (GCI) (World Economic Forum, 2019) released by the World Economic Forum and the Global Knowledge Economy Index (KEI) released by the World Bank have formed internationally influential evaluation systems for innovation capability in foreign countries (World Bank, 2007).

With the comprehensive implementation of the innovation-driven development strategy in China, the measurement of innovation-driven development levels has attracted extensive attention from the academic circles, and a series of research results have been achieved at national, regional, provincial and municipal levels. Zhou Ke, Tang Juanli, and Gu Zhouyang (2018) constructed an evaluation system consisting of four first-level indicators (innovation foundation/conditions, innovation input, innovation output, and innovation contribution/impacts), 14 second-level indicators, and 39 third-level indicators, and conducted a comprehensive evaluation of China’s innovation-driven development capability using the entropy weight method. Shao Hanhua and Qi Rong (2019) constructed an innovation-driven development index system applicable to cities and evaluated and analyzed regional differences and dynamic evolution of innovation-driven development levels in the Yangtze River Economic Belt from three perspectives: innovation input and output, regional innovation environments, and innovation-driven effects. Li Xuhui, Chen Ying, and Cheng Gang (2020) constructed an evaluation index system for innovationdriven development of the Yangtze River Economic Belt under the “Driver-Pressure-State-Impact-Response” (DPSIR) framework. This system contains 31 specific indicators in five criteria layers, including innovation drivers, innovation pressures, innovation states, innovation impacts, and innovation responses. Yin Mengji (2015) believes that the output of scientific and technological innovation in a region is mainly reflected in the production and sales of new products, the number of new patents, and the transaction amounts in the technology market. He measured the economic development using GDP and the technological innovation levels using the sales revenue of new products, the number of patents approved, and the transaction amounts in the technology market, and made an empirical study of the innovation development levels of provincial administrative regions in China. Liu Zuojing, Yan Xiaoxu, and Chen Jianxin (2018) constructed an evaluation model integrating regional innovation and economic development based on the coupling theory. The regional innovation system consists of three indicators, including innovation input, innovation output, and innovation environments, as does the economic system, including the economic aggregate, economic structure, and economic benefits. Their paper measured and evaluated the coupling coordination degree of the innovation and economic systems of Guangdong province from spatial and temporal perspectives. Yuan Yong and Hu Haipeng (2020), based on the connotation of innovation-driven development, established a comprehensive evaluation system for measuring innovation-driven development at the municipal level from five aspects, including the provisions of an innovation environment, gathering of innovation resources, allocations of innovation resources, innovation outputs, and innovation’s effects on economic and social development. In general, Chinese academics tend to construct the system for evaluation of innovation-driven development from three perspectives: by selecting several representative indexes based on the “input-output” model; by considering such factors as innovation environments, innovation resources, and innovation subjects based on the “inputoutput” model by applying the DPSIR model. At present, most academics construct the evaluation systems based on the “input-output” model, and they tend to consider more factors such as innovation factors, innovation environments, and environmental subjects in their research instead of merely considering innovation input and output. In addition, although the evaluation systems built based on the DPSIR model consider the interactions between different factors, simple linear causality has excessively simplified the actual process of index selection (Cao, 2005). Therefore, less representative and low-causality factors may be selected, leading to inaccurate results. To this end, we developed an evaluation system that can reflect the process and results of innovationdriven development by fully considering factors such as innovation factors, innovation subjects, and achievement transformations based on the “input-output” model. In terms of research methods, domestic research results are mainly achieved through the entropy weight method. Supported by ample data, this method can eliminate overlapped information among various index variables and effectively avoid the errors caused by human (subjective) factors, acquiring highly objective results. Moreover, it has been verified by various empirical studies. Therefore, drawing on the existing results, we adopted the entropy weight method to measure the innovation-driven development of Sichuan province, to ensure the objectivity of the evaluation results.

Evaluation System, Research Methodology, and Data Sources

Evaluation System

Innovation-driven development involves not only the gathering of innovation factors, the allocation of innovation resources, and the output of innovation results but also the application and diffusion of innovation results and innovation-driven economic and social development. Therefore, the evaluation of the innovation-driven development levels focuses on both the driving role of innovation in economic and social development and the link between the front end and the back end of innovation-driven development (Yuan & Hu 2020). Based on the connotations and various links of innovation-driven development and the principles of scientificalness, systematicness, representativeness, and accessibility, we developed an evaluation index system by reviewing relevant literature and considering the regional development characteristics of Sichuan province through repeated screenings, adjustments, and optimizations. The evaluation index system, taking the comprehensive index of innovationdriven urban development as the target layer, consists of 20 indexes with five categories each, including innovation factors, innovation subjects, innovation environments, innovation outputs, and development performances.

(a) Innovation factors: This index mainly reflects the allocation of innovation factors such as human, capital, and material resources. Innovation factors are at the front-end of innovationdriven development and the basis for innovation subjects to conduct innovation activities. They are mainly reflected in fund investments, talent support, and platform construction.

(b) Innovation subjects: Innovation subjects are the main force in conducting scientific and technological innovation, mainly including enterprises, colleges and universities, scientific research institutes, industrial technology institutes, and other new R&D institutions.

(c) Innovation environments: This mainly refers to the external environments of innovationdriven development. A good innovation environment can provide conditions and support for the implementation of innovation activities and can guide and promote scientific and technological innovation by the innovation subjects. It is mainly reflected by relevant indicators such as policy environments, market-oriented levels, and construction of public facilities.

(d) Innovation output: It mainly refers to the achievements attained in the innovation process. Knowledge is the source of innovation and is created by recombining and reallocating innovation resources to promote the development of high-tech industries through the transformation of scientific and technological achievements, thus forming new markets and economic growth points. Innovation output is mainly reflected through relevant indicators such as knowledge creation, achievement transformation, and industrial development.

(e) Development performance: This mainly refers to the level of innovation-driven economic and social development, i.e., to what extent are economic development and social progress promoted through the application of innovation output in economic and social activities. It is mainly reflected by economic efficiency, production efficiency, people’s livelihoods, environments, and other related indexes.

Research Methodology

We determined the weight of each index in the index system for evaluation of innovationdriven development capability using the entropy weight method and then measured and horizontally compared the comprehensive indexes of urban innovation-driven development based on the weighted sum model. In information theory, entropy is defined as a measure of the disorder of the system. It has been widely used in many fields, such as the evaluation of innovation-driven development levels and sustainable development capacities. Based on the values of each index, the weight of indexes was determined using the entropy weight method according to their dispersion degrees. The higher the dispersion degree of an index, the lower the entropy. This means this index has a greater degree of differentiation and carries more information, thus having a greater influence on the comprehensive evaluation of each factor in the system. The entropy weight method is an objective weighting method that can effectively measure the importance of indexes in the evaluation system based merely on the differentiation between the values (He, Xie & Wang, 2020). The specific steps are as follows (Zhou, Tang & Gu, 2018; Zhu & Wei, 2015):

(a) Nondimensionalize the data. Since the evaluation system has many indexes that are measured in different units, it is impossible to proceed with the calculations directly. Therefore, the basic data needs to be nondimensionalized to eliminate the impacts caused by the different dimensions of the various indexes.

where,X'ijrefers to the nondimensionalized value of indexjof Cityi;Xijrefers to the original value of indexjof Cityi;is the average value of indexj; andSjis the sample standard deviation of indexj.

Calculations using Formula (1) may generate some negative dimensionless data. However, since the data used in the entropy weight method must be positive, such negative data needs to be processed through translation to obtain new data.

where,Aijis the value after translation ofX'ijand a is the translation amplitude. To minimize the error factor in the basic data, caused by translation, and ensure more obvious evaluation results, the value ofashould be as close to ｜min(X'ij)｜ as possible.

(b) Calculate the proportionPijof the valueAijafter translation in the indexjof thencities.

where,nis the number of samples (cities and prefectures) andmis the number of evaluation indexes.

(c) Calculate the information entropyEjof indexj.

where,Kis a constant,

(d) Calculate the value of information utilityGiof indexj. The value of information utility indicates the effects of the index on the research object. The greater the value of information utility, the greater the weight of the index.

(e) Calculate the weightWiof the indexj.

(f) Calculate the comprehensive indexCijof innovation-driven urban development.

Data Sources

The data in the evaluation index system selected for this study came from the Sichuan Statistical Yearbooks and the statistical yearbooks of the various cities and prefectures for the relevant years. Therefore, all the data sources cited in this paper are authentic and dependable. To ensure the availability and comparability of the data, the “science and technology innovation platforms” mentioned in this paper mainly include engineering research centers, key laboratories, and enterprise technology centers at the provincial level and above, and the “number of high-tech business incubators” mainly refers to the high-tech business incubators registered at the national and provincial levels. The “proportion of output value of the high-tech industry in operating income of industrial enterprises” is derived from the data of industrial enterprises with a scale above the designated one, and the “market-oriented level” is reflected by the “proportion of the added value of the private economy in regional GDP.”

Evaluation Process, Results, and Analysis

Determination of Index Weights

The data in the 20 indexes of the 21 cities and prefectures of Sichuan province in 2019 were processed using formulas (1) – (6) to reflect the weight of each evaluation index. Based on the results, the weights of the indexes in the criteria layer are calculated. As noted in Table 2, the innovation environment has the greatest weight in the criteria layer, accounting for 23.42%, followed by development performance with a weight of 22.02%, and innovation output with a weight of 19.16%. This indicates that a good innovation environment is a key factor in improving the innovation-driven development level of the cities and prefectures in Sichuan province. Development performance is the ultimate goal of innovation-driven development, which has a significant impact on the innovation-driven development levels of the 21 cities and prefectures. It is also an effective way to improve the innovation-driven development levels by applying innovation output such as created knowledge and achieved results to industrial development and social life to drive economic and social development. In terms of specific indexes, the market-oriented level accounts for the highest proportion, reaching 8.74%, indicating that an improvement in the market-oriented level has a very positive impact on the technological innovation of enterprises.

Table 1 Index System for Evaluation of Innovation-Driven Urban Development Capability in Sichuan Province

Table 2 Weights of Evaluation Indexes

Calculation of Combined Scores

Calculations using Formula (7) reveals the combined scores and sub-scores for innovationdriven development levels for the 21 cities and prefectures of Sichuan province in 2019. These scores are presented in Table 3, and there are obvious spatial differences in innovation-driven development levels of the 21 cities and prefectures in Sichuan province. Chengdu gets the highest combined score, which is 1.5 times that of Mianyang, ranking second, and nearly 3 times that of Ganzi Tibetan Autonomous Prefecture, ranking last.

Table 3 Combined Scores for Innovation-Driven Development Levels of Sichuan Province in 2019

Analysis of Evaluation Results

According to the combined scores of the innovation-driven development levels, the 21 cities and prefectures of Sichuan province were classified into four categories: advanced-level area (with a combined score higher than or equal to 0.1000), high-level area (with a combined score higher than or equal to 0.0500 but lower than 0.1000), medium-level area (with a combined score higher than or equal to 0.0400 but lower than 0.0500), and low-level area (with a combined score lower than 0.0400).

The advanced-level area includes only Chengdu, the capital of Sichuan province, with a combined score of 0.1000, far exceeding the other cities in Sichuan province. Chengdu is the economic, political, and cultural center of Sichuan province and the comprehensive national science center approved by the state. Chengdu boasts a booming high-tech industry, gathering a variety of innovation factors and subjects such as scientific and technological talents, colleges and universities, scientific research institutions, and high-tech enterprises. With a strong scientific and technological foundation and innovation strength, Chengdu shows absolute advantages in terms of innovation-driven development levels and classification indexes such as innovation factors, innovation subjects, innovation environments, innovation outputs, and development performances. As a city that comes first in terms of the innovation-driven development levels of Sichuan province, Chengdu has played a certain demonstration role in the innovation-driven development of all areas of the province, but its role as a driving force is far from sufficient.

The high-level area includes Mianyang, Panzhihua, and Deyang, with combined scores of 0.0654, 0.0567, and 0.0527 respectively. Mianyang is the only science and technology city in China. Owing to policy advantages in gathering innovation resources and cultivating innovation subjects, Mianyang has effectively improved its overall innovation-driven development level. However, its efficiency in transforming scientific and technological achievements is relatively low, leading to limited effects in the actual application of innovation output and a relatively weak development performance. Based on its distinctive industry of vanadium and titanium, Panzhihua has gathered a number of innovation platforms, incubators, and scientific and technological talents in the field of vanadium and titanium. It has made remarkable achievements in innovation factors, innovation subjects, innovation outputs, and development performances, but its innovation environment needs to be further improved. Focusing on promoting the construction of the Deyang National High-Tech Industrial Development Zone, Deyang is gathering innovation factors, cultivating innovation subjects such as high-tech enterprises, and accelerating the development of high-tech industries, with certain advantages in the whole province in terms of its innovation-driven development level.

The medium-level area includes 13 cities and prefectures such as Zigong, Yibin, Luzhou, Ya’an and Leshan, accounting for more than 60% of the whole province. They are mainly distributed in the Southern Sichuan Economic Zone, Northeast Sichuan Economic Zone, and Chengdu Plain Economic Zone, with certain differences between them. Zigong has certain advantages in innovation factors, innovation outputs, and development performances, but its innovation environment needs to be further optimized. Compared with its ranking in terms of its innovation-driven development level, Yibin falls behind in terms of innovation environments and innovation outputs. The innovation output of cities such as Neijiang, Nanchong, and Guangyuan needs to be further improved. Suining and Nanchong have made remarkable achievements in cultivating innovation subjects and creating innovation environments. Ya’an performs excellently in terms of the innovation environments and innovation factors. However, due to the lack of innovation subjects, it lags behind as far as innovation outputs and development performances are concerned. In general, the innovation-driven development levels of these areas is well-matched with their overall strength. Cities in the Southern Sichuan Economic Zone such as Zigong, Yibin, and Luzhou are generally superior to those in the Northeast Sichuan Economic Zone such as Nanchong, Dazhou and Guangyuan in terms of innovation-driven development levels. Cities in the Chengdu Plain Economic Zone such as Ya’an, Leshan, and Suining vary greatly when it comes to the innovation-driven development levels. Ya’an has a combined score of 0.0455, ranking 8th, while Ziyang also in Chengdu Plain Economic Zone, ranks 17th with a combined score of 0.0407.

The low-level area includes Bazhong, the Aba Tibetan and Qiang Autonomous Prefecture, the Liangshan Yi Autonomous Prefecture, and the Ganzi Tibetan Autonomous Prefecture, each with a combined score lower than 0.0400. Bazhong hits a relatively high score in respect of the innovation environments. However, restricted by other factors like its geographical location and industrial foundations, it performs poorly in gathering innovation factors and gets a relatively low score in innovation subjects, innovation outputs, and development performances. The Liangshan Yi Autonomous Prefecture, the Aba Tibetan and Qiang Autonomous Prefecture, and the Ganzi Tibetan Autonomous Prefecture are three minority autonomous prefectures populated by ethnic minorities in Sichuan province. The Ganzi Tibetan Autonomous Prefecture and the Aba Tibetan and Qiang Autonomous Prefecture are within the Northwest Sichuan Ecological and Economic Zones, with vulnerable ecological environments and weak economic strength. These two prefectures are positioned to focus on protecting their ecological environments and developing the tourism industry by relying on rich local tourism resources. Therefore, there is a limited demand for innovation, leading to a low innovation-driven development level. The Liangshan Yi Autonomous Prefecture is partly in the Panxi Pilot Zone of Innovation and Development of Strategic Resources, where scientific and technological innovations are made centering on the innovation-driven development of vanadium and titanium resources. The Liangshan Yi Autonomous Prefecture has made certain achievements in cultivating innovation subjects and boosting innovation outputs. Although it occupies a large area, due to its location in an area of ethnic minorities with a vulnerable ecological environment, the Liangshan Yi Autonomous Prefecture has seen limited development in some areas and has obvious disadvantages in terms of the innovation environments and development performances, resulting in an overall low innovation-driven development level.

Conclusions and Discussions

We developed a comprehensive evaluation index system to measure the innovation-driven development levels in Sichuan province. This system contains 20 indexes with five aspects each, including innovation factors, innovation subjects, innovation environments, innovation outputs, and development performances. The entropy weight method is adopted to make a comprehensive evaluation of the innovation-driven development levels of 21 cities and prefectures of Sichuan province in 2019. According to the evaluation results, the spatial patterns of innovation-driven development in Sichuan province were analyzed and the following conclusions were drawn.

(a) Innovation-driven development runs through the whole process of innovation from carrying out innovation activities and making scientific and technological achievements, to applying achievements and driving development. The evaluation system built to measure the innovation-driven development levels in Sichuan province is composed of a variety of indexes. We used the entropy weight method to determine the weight of each index as this method can effectively eliminate the overlapping information among indexes and more objectively reflect the importance of different indexes in the evaluation system.

(b) Innovation environments and development performances are the major indexes that affect the innovation-driven development of Sichuan province, followed by innovation outputs, innovation factors, and innovation subjects. The market-oriented level is the most important index that affects the innovation-driven development levels, which means it is crucial to promoting market-oriented reform and creating a good innovation environment under the innovation-driven development strategy. Moreover, while gathering innovation factors, cultivating innovation subjects, and increasing innovation output, attention should be paid to enhancing the ability to transform and apply innovation achievements, and to accelerating the application of scientific and technological achievements in economic and social development to make innovation the primary driving force for overall economic and social development.

(c) According to the comprehensive evaluation results, Sichuan province is divided into areas defined by the four categories of innovation-driven development levels: the advancedlevel area including Chengdu only; the high-level area, including Mianyang, Panzhihua, and Deyang; the medium-level area, including 13 cities and prefectures such as Zigong, Yibin, Luzhou, Ya’an and Leshan, accounting for more than 60% of the whole province; and the low-level area including Bazhong, the Aba Tibetan and Qiang Autonomous Prefecture, the Liangshan Yi Autonomous Prefecture, and the Ganzi Tibetan Autonomous Prefecture.

(d) There are obvious spatial differences in the innovation-driven development levels in Sichuan province. In general, the Chengdu Plain Economic Zone, with Chengdu, Deyang, and Mianyang as the core, ranks first in the innovation-driven development levels. However, there are relatively large internal differences between these three cities in this aspect, i.e., Chengdu is at an advanced level, Deyang is at a high level, and Mianyang is at a medium level. This is also true in the Panxi Pilot Zone of Innovation and Development of Strategic Resources. Within the Zone, Panzhihua ranks third and the Liangshan Yi Autonomous Prefecture ranks second to last, while other cities are at the medium and low levels. The cities in the Southern Sichuan Economic Zone are generally superior to those in the Northeast Sichuan Economic Zone in terms of innovation-driven development levels. This is to be expected because some cities and prefectures are affected by natural conditions, industrial foundations, and other factors, and obvious differences are observed in their innovation environments, innovation outputs, innovation factors, and innovation subjects.

(e) Although other studies have been conducted on the evaluation of innovation-driven development levels in the academic circles, no consensus has yet been reached. The evaluation index system we constructed, and presented here, considers the availability of data and the comparability between regions, but there are still constraints in selecting indexes and collecting data. For example, the policy environment under the index of innovation environments is only reflected by the “additional tax deductions for expenditure on R&D.” Indexes such as scientific and technological innovation platforms and scientific and technological enterprise incubators only apply to institutions above the provincial level; the market-oriented levels are represented by a single index of “proportion of the added value of private economy in regional GDP.” Therefore, we are unable to comprehensively reflect the actual situations in terms of innovation factors and innovation environments for each city and prefecture of Sichuan province, which may have an impact on the evaluation results and thus needs to be further verified through future research.