Abstract
The purpose of this article is threefold. First, we measure both technical and scale efficiency of major Australian banks during the period 2006 to 2012 using data envelopment analysis (DEA). Second, the robustness and sensitivity of efficiency scores are tested based on a bootstrap approach. Third, the bootstrap results are employed to examine by how much different scenarios of improving technical efficiency scores can result in cost savings for individual banks. The results indicate a notable difference between the original and bootstrap efficiency estimates. While the conventional DEA results show that more than half of banks are fully technical efficient, the bootstrap results reveal a more accurate account of efficiency disparities from best-practicing banks.