AN EMPIRICAL INVESTIGATION OF COMMERCIAL BANKS’ EFFICIENCY IN PAKISTAN: A NON PARAMETRIC DATA ENVELOPMENT APPROACH

Abstract

The objective of this study was to examine technical (TE), pure technical (PTE) and scale (SE) efficiencies of 32 commercial banks (CB) in Pakistan for the year 2009. The study applied a non-parametric Data Envelopment Analysis (DEA) to measure efficiencies. For the selection of inputs and outputs variables, an intermediation approach was adopted. These variables were Number of employees, Physical Capital and Borrowed funds, as inputs while net advances, investments and lending to financial institutions as outputs. Two basic models of DEA namely CCR and BCC were applied. The empirical results of these models disclosed that CB functioned at 93 percent level of TE which means that CB can achieve the same level of outputs by using 7% less than current inputs used. The technical inefficiency was mainly caused by scale size (4%) than managerial inefficiency (3%). The pre-dominate cause of scale inefficiency was observed to be decreasing return to scale. The most efficient (highly robust) banks were Bank of Punjab, Habib Metropolitan, Muslim Commercial bank, Burj Bank, and Faysal Bank while Bank Islami and HSBC Oman were indicated as most inefficient banks.