关键词: 高新技术产业, 创新效率, 随机前沿生产函数模型

Abstract: Innovation efficiency is the key factor affecting the division of the labor status of the high-tech industry in the global value chain. The innovation ability of the high-tech industry is an important basis for the comprehensive competitiveness of a country or region. In recent years, China’s high-tech industry has developed rapidly, but it is also faced with many problems, such as the low intensity of R&D investment, the weak ability of independent innovation, fragmentation of innovation resources, imperfect industrial collaborative innovation system and so on. All these are related to the low efficiency of innovation in high-tech industries. Overall, the areas to be expanded in this field include: firstly, bringing the intermediate output and final output of high-tech industries into a unified analytical framework. The differences in efficiency values produced by different output indicators and the reasons for the differences were compared and analyzed. Secondly, in-depth analysis of the spatial relevance and regional differences of the innovation efficiency of the high-tech industry, and then explore the effective ways to realize the regional balanced and coordinated development of the high-tech industry.
The technological innovation of high-tech industry is a process of the joint action of endogenous factors and exogenous factors, so it is necessary to fully consider the influence of exogenous variables to measure its efficiency. In the process of estimation, the maximum likelihood method ensures that the estimated sample information can be used equally and fully. Therefore, this paper chooses to use the SFA method and uses the relevant data of 27 provinces or cities in China from 2000 to 2015, which analyzed the spatio-temporal differentiation characteristics and influencing factors of innovation efficiency in China.
The results show that: (1) The overall level of high-tech industry’s innovation efficiency is not high, especially the innovation efficiency with new product sales income as the final output is the most obvious, and the overall average level of innovation efficiency of regional high-tech industry is low. It was only 0.282, and the inefficiency factor accounted for 0.718. Taking the effective number of patents in the high-tech industry, the result shows that the overall average level of innovation efficiency is 0.781, and the inefficiency factor accounts for 0.219. (2) The innovation efficiency of the high-tech industry among regions generally shows a fluctuating increase, and the regional distribution of its growth rate is different. The variation range of the overall innovation efficiency of a high-tech industry estimated is \[0.08,0.48\] by new products. However, the variation range of the overall efficiency of the high-tech industry estimated is \[0.710.85\] by the effective number of patents, and there is no obvious upward trend. (3) In the innovation efficiency of high-tech industry, the efficiency of technology loss mainly comes from the degree of opening to the outside world and government support, among which the degree of opening to the outside world accelerates the loss of innovation efficiency of the high-tech industry. On the other hand, government support delays the loss of innovation efficiency. (4) Under different conditions, the effects of R&D funds stock and R&D human capital stock on the innovation output of the high-tech industry are not the same. (5) From the patent as an intermediate product to the sale of new products as final products, the innovation efficiency of high-tech industry loses a lot in this intermediate process. (6) The regions with average annual growth in innovation efficiency of high-tech industries in various provinces include the northeast (0.856%), the eastern coastal areas (0.841%), the northwest (0.576%), the northern coastal areas (0.486%), the southern coastal areas (0.207%) and the middle reaches of the Yangtze River (0.044%), On the other hand, the areas where the average annual growth rate of innovation efficiency of high-tech industries decreases are the middle reaches of the Yellow River (-0.759%) and the southwest (-0.006%). Comparatively speaking, there is a large gap in innovation efficiency among different regions, especially between the eastern region and the southwest-middle reaches of the Yellow River. The innovation efficiency of the adjacent areas in the east has increased, which shows that the knowledge spillover effect in the east is obvious.
The policy implications of this study includes: firstly, optimizing the development environment of high-tech industry, focusing on strengthening the protection of intellectual property rights, perfecting the system and mechanism of technological achievement transfer and transformation in high-tech industry, Expand the industrial chain between intermediate output and final output of high-tech industry, and promote the organic integration of innovation chain, entrepreneurial chain, capital chain, talent chain and policy chain. Secondly, we should strengthen the investment in innovation, especially encourage high-tech enterprises to increase the investment in R&D, which can be supported by implementing and perfecting the policy of deducting R&D expenses. Thirdly, to formulate differentiated policies for the development of high-tech industries, for a long time to come, it is necessary to give appropriate policies to the backward areas in the development of high-tech industries, and at the same time establish and improve the regional linkage mechanism. To promote the coordinated development of the industry; Fourth, we should improve the openness of high-tech industries, attract more foreign direct investment into the field of high-tech industries, and encourage more high-tech enterprises to carry out transnational mergers and reorganization, transnational technology transfer and transnational investment.

Key words: high-tech industry, innovation efficiency, stochastic frontier production function model