在采用TOPSIS对科技产出进行评价的基础上,利用DEA分析了研发人员全时当量、银行贷款、政府与企业科技投入对科技产出的效率,并采用生产函数和面板数据对有无效率损失情况下投入要素的贡献进行了比较。研究发现,中国科技投入产出效率总体不高,经济发展水平对科技投入产出效率的影响不大。从投入要素的利用效率看,政府科技投入的利用效率最高,企业科技投入的利用效率第二,研发人员全时当量的利用效率第三,银行信贷的利用效率最低。由于效率损失,导致政府科技投入的贡献被高估,而企业科技投入的贡献被低估了。科技贷款的利用效率最低,并且弹性系数为负数,深层次暴露了科技金融在信用体系和风险体系方面存在的问题。
This paper analyzes science and technology efficiency based on R&D personnel FTE, loan, the government and enter- prise input after evaluating S&T output with TOPSIS, then compares the contribution of the input factors under efficiency and non-effi- ciency based on panel data and product function. The results show the efficiency of China S&T input-output is low, but the economy development level has little influence on efficiency of S&T input-output. The input factors ranking in descending order of utilization effi- ciency is government, enterprise, R&D personnel FI'E, and loan. Due to efficiency loss, the contribution of the government S&T input is over-estimated while that of the enterprise S&T input is under-estimated. The elastic ratio of loan S&T input is negative and its effi- ciency is the lowest. Therefore, problems of S&T finance in credit system and risk system are exposed in a deep manner.