Empirical study of innovation efficiency of China's high-tech industries using two-stage stochastic super-efficiency nonparametric technologies

High-tech industry becomes China’s leading industry due to its knowledge-intensive and substantial value-added characteristics. It is important to note that innovation efficiency of the high-tech industry directly affects its development level. Empirical study on innovation efficiency of the high-tech industry can provide meaningful insights into their high-quality development. In this respect, this paper first regards the high-tech industry as a two-stage chain structure and then develops two-stage stochastic super-efficiency nonparametric technologies, focusing on the combination of two-stage super-efficiency methods with stochastic properties of indicators, to perform the efficiency evaluation of the high-tech industry. Empirical findings indicate that neglecting intermediate elements of the high-tech industry can lead to biased results. Moreover, the super-efficiency model may enhance the ability to completely differentiate the development level of the high-tech industry across provinces. In addition, nonconvex technology can better reveal development differences across sub-stages within the industry. Finally, based on the result analyses, some constructive suggestions are provided for policymakers to create an environment conducive to industrial development.