Bank branch efficiency under environmental change: A bootstrap DEA on monthly profit and loss accounting statements of Greek retail branches

The objective of this study is to measure efficiency change of bank branches under external environment deterioration. In particular, we utilize a bootstrap input-oriented profit DEA and investigate homogeneous and heterogeneous branches according to branch size and location to measure efficiency change by contrasting expansion, recession and capital control effects that constitute a unique phenomenon in the postwar period in the Eurozone. Our primary research explicitly focuses on the whole retail network of a Greek systemic bank based on unpublished monthly branch Profit and Loss statements and covers the period from January 2006 to July 2016. We find that early and deep recession reduces on average branch network efficiency. The imposition of capital controls (end-month June 2015) initially causes marginal effects with a subsequent efficiency improvement in the first seven months of 2016 when economic conditions are normalized. The paper documents that branch size and location matter. On the whole, we capture efficiency deterioration in the long-run contrary to recent European evidence. Apart from the efficiency measurement over time, we provide directions to bank management for performance improvement in the capital control period. More specifically, a bootstrap DEA-based Decision Tree classification exactly quantifies for the first time a potential upgrading of underperforming branches and a second-stage bootstrap DEA regression locates important efficiency drivers such as the diversification of income and the deposit- oriented activity that could improve efficiency of the total retail network.