Aggregate efficiency of industry and its groups: the case of Queensland public hospitals

Abstract

In this paper, we explore the efficiency of different groups of hospitals in Queensland, Australia, focusing on teaching and non-teaching hospitals, by adapting the most recent developments on statistical analysis of aggregate efficiency. We focus on the two approaches: the bootstrap approach proposed by Simar and Zelenyuk (J Appl Econ 22(7):1367–1394, 2007) and the central limit theorems recently developed by Simar and Zelenyuk (Oper Res 66(1):137–149, 2018), (Eur J Oper Res, 2020). To adapt these developments, we extend the central limit theorems to the context where there are several sub-groups in the population. Using real data on Queensland public hospitals, we found that teaching hospitals are significantly less efficient than non-teaching hospitals when benchmarking is done with respect to the constant returns to scale frontier, but are significantly more efficient when benchmarking with respect to the variable returns to scale frontier.