中国创新型产业集群投入产出效率动态演进及区域差异——基于省际面板数据的分析

基于2014—2019年中国内地28个省域面板数据，构建三阶段DEA模型，测度各省域创新型产业集群投入产出效率，运用变异系数和静态面板模型分别检验其σ收敛和β收敛特征，进而揭示东部、中部、西部、东北地区四大区域创新型产业集群投入产出效率的动态演进趋势及区域差异。结果表明，2014—2019年全国创新型产业集群投入产出效率由0.62提高至0.72，更多省域向高效率演进，到2019年，35.7%的省域达到生产效率前沿；中国创新型产业集群投入产出效率存在明显的区域差异，在空间上呈现出东部>中部（与全国平均水平持平）>东北和西部的发展格局，较低的纯技术效率、规模效率分别是制约东北和西部地区效率提升的主因；样本考察期内，全国整体、东部和中部创新型产业集群投入产出效率存在σ收敛，全国整体及东部、东北和西部地区存在绝对β收敛和条件β收敛，而中部仅存在条件β收敛，且不同区域条件β收敛的影响因素不尽相同。据此，针对提升不同区域创新型产业集群投入产出效率提出针对性建议。

Dynamic Evolution and Regional Differences of the Input-output Efficiency of China's Innovative Industrial Clusters:An Analysis Based on Provincial Panel Data

In China, the construction of innovative industrial clusters is a policy attempting to break through the low-end lock-in of traditional industrial clusters and promote industrial upgrading. Since the Ministry of Science and Technology of China officially launched the pilot work of innovative industrial clusters in 2013, all localities have achieved remarkable results in cluster construction. However, the development of innovative industrial clusters in various regions of China is uneven. Although a few clusters have ranked among the top 100 global innovation clusters, most clusters have low production efficiency, and some are even up against the challenge of survival. In addition, the problem of unbalanced resource allocation has increasingly hindered the coordinated development of regional innovative industrial clusters. The above problems not only restrict the overall high-quality development of China's innovative industrial clusters, but also hinder the improvement of regional innovation capabilities and industrial competitiveness. This paper argues that for both the whole country and each region, only by measuring the input-output efficiency of innovative industrial clusters objectively can we use limited resources to achieve more output. However, the spatial distribution characteristics and time-series evolution of the input-output efficiency of China's innovative industrial clusters are still unclear, and it is still impossible to accurately predict its development trend.#br#Therefore, based on the panel data of 28 provinces in China from 2014 to 2019, this paper constructs a three-stage DEA method to measure the input-output efficiency of innovative industrial clusters in each province, and uses the coefficient of variation and static panel model to analyze their σ convergence and β convergence characteristics, respectively, thus revealing the dynamic evolution trend and regional differences of the input-output efficiency of innovative industrial clusters in east, central, northeast and west regions of China. This study enriches the literature on the evaluation of the input-output efficiency of industrial clusters. In practice, the conclusions of this paper are helpful for the Chinese government to judge the basic situation and development trend of the input-output efficiency of innovative industrial clusters as a whole, identify regional efficiency differences and their underlying causes, and thus providing reference for decision-making departments to formulate relevant policies.#br#The results show that the input-output efficiency of innovative industrial clusters was increased from 0.62 in 2014 to 0.72 in 2019 on the national level, and more provinces were evolved to high efficiency. By 2019, 35.7% of the provinces have reached the frontier of production efficiency. There are obvious regional differences in the input-output efficiency of China's innovative industrial clusters, which shows a spatial development pattern that the eastern is higher than the central (the same as the national average approximately) and higher than the northeast and western regions. Low pure technical efficiency and scale efficiency are the main factors restricting the efficiency improvement in the northeast and west regions, respectively. During the sample investigation period, the input-output efficiency of innovative industrial clusters in the whole country, the eastern and central regions have σ convergence, and there are absolute β convergence and conditional β convergence in the whole country, eastern, northeast and western regions, while there is only conditional β convergence in the central regions, and the influencing factors of conditional β convergence are different in different regions.#br#Based on the above conclusions, this paper puts forward the following policy recommendations. First of all, There is still much room to improve the input-output efficiency of China's innovative industrial clusters, and local governments should clearly recognize the importance and urgency of promoting the input-output efficiency of innovation-oriented industrial clusters. Secondly, the development of innovative industrial clusters should be coordinated on the national level, and differentiated and targeted policy measures should be adopted according to the actual situation of the input and output of the clusters in various regions to promote balanced regional development. Finally, in the process of promoting the balanced development strategy of input-output efficiency of innovation-oriented industrial clusters, we should not only keep narrowing the input-output efficiency gap between regions and provinces, but also take into account the coordination of efficiency improvement speed between regions.#br#