The Efficiency of Futures Research Institutes of China
From the perspective of comprehensive efficiency, the 40 futures companies scored a maximum of 1, a minimum of 0.095 and an average of 0.395, which shows that there is a large gap in efficiency between the research institutes of futures companies. Specifically, the DEA reveals 3 companies with the best efficiency, namely, Jinshi Futures, Dayou Futures and BOC International Futures, which account for 7.5% of the research objects. These three companies are also effective in pure technical efficiency and scale efficiency, which shows that they have the right input-output ratios for their research institutes, have fully utilized the labour and costs invested and have achieved the maximum output results at this input level. There are 37 companies that are not efficient according to the DEA, accounting for 92.5% of the sample. Among them, 5 companies that are slightly inefficient and 32 companies that are seriously inefficient, accounting for 12.5% and 80% of the sample, respectively. This shows that the efficiency level of the research institutes of futures companies is generally low, and a large number of research institutes have not achieved comprehensive efficiency and effectiveness. From the perspective of pure technical efficiency, the maximum value is 1, the minimum value is 0.099, and the average value is 0.468. There are 7 company research institutes with efficient technical efficiency, accounting for 17.5% of the sample. These companies have invested reasonable resources in their research departments. They are inefficient in terms of comprehensive efficiency but efficient in pure technical efficiency, which means that the main reason that the DEA scores them as inefficient their inefficiency in scale efficiency. There are 33 research institutes with poor technical efficiency, accounting for 82.5% of the sample and indicating that the allocation of research department resources by futures companies requires further improvement. From the perspective of scale efficiency, the maximum value is 1, the minimum value is 0.356, and the average value is 0.876. All decision-making units with poor DEA evaluations also have poor scale efficiency. The inefficient comprehensive efficiency of research departments is attributable to pure technical efficiency and scale efficiency. In terms of returns to scale, 14 research institutes have constant returns to scale, accounting for 35% of the sample, and the returns are economic. There are 7 futures companies with increasing returns to scale, accounting for 17.5% of the sample. There are 19 companies with diminishing returns to scale, accounting for 47.5% of the sample.