Bank efficiency in Bangladesh revisited: a slack-based network DEA approach

Purpose Data envelopment analysis (DEA) calculates the efficiency of a business unit if all the inputs are creating outputs within a “black box.” Under traditional DEA, the detailed process of that business unit is ignored. However, a network DEA can explain the black box structure and provide efficiency results for sub-sections within any business process. This study aims to propose a network DEA model that explains a bank's total operation. Design/methodology/approach Earlier studies have focused only on bank efficiency ignoring this breakdown. This study departs from them by using a slack-based two-stage network DEA under a novel banking business perspective. Findings The results reveal that network DEA provides better benchmarking insights than the traditional DEA. As such, better benchmarking can guide both the banking industry managers and policy makers in Bangladesh. Originality/value The major contribution of this study includes dividing a bank's total operation efficiency into two sub-operations: “core operations – collecting deposits and giving loans” and “additional operations – fees, commissions and other services.”