Data Envelopment Analysis
This paper applies DEA to analyze the x efficiency of the Pakistani banking sector. DEA assumes that all underlying units use the similar amount of technology to make output by using the available input. In DEA, efficiency in production is determined by using Decision Making Units (DMUs) which help examiner to identify the best performing unit and to compare its performance with other units. This technique enables the evaluator to use multiple outputs and inputs. Under this technique the score assigned to DMUs range from 0 to 1, whereas the 1 is given to the most efficient. It is not necessary to gain score of 1 for best practicing or most efficient unit or to give maximum output. A best Decision making unit is one which provides maximum output from given set of input. Hence DEA is a preferred technique for measuring efficiency even dealing with small sample size (Isik and Hassan, 2002 and Ataullah et al., in 2004).
Two ways to select inputs and outputs to measure bank’s efficiency are; operating and intermediating approach. In first approach bank is considered to be provider and producer of different services for its clients and is evaluated form cost/revenue administration perspective. The intermediation method is given by, Sealey and Lindley, which completes the operating approach. In second approach bank is considered to be an intermediary of financial services and convert these resources into loans and other assets. This method provides various benefits as compared to production approach for analyzing financial institutions as it includes interest and/or funding expenses.
Many DEA models have been used to measure the efficiency. Widely used models are developed by Charnes, Cooper and Rhoades (CCR) model of Charnes et al. and Banker, Charnes and Cooper (BCC) model of Banker et al.. CCR takes the firm or DMUs as function of CRS (constant return to scale) whereas BCC extends the model to variable return to scale.
Equation ensures efficiency ratio to be at least one and the equation ensures that weights are positive.
Specification of Inputs and Outputs
As Isl. banks follow equity based capital structure, so this paper used intermediation approach to determine efficiency with 3 outputs and 3 inputs.
X1: Investment and Financing X2: Total Income X3: Liquid Assets
y1: Administrative Costs y2: Operating Fixed Assets y3: Total Deposits
Commercial banks are used in the study to conduct analysis. Only those banks are selected for which data is available for the period 2006-2010. Following this criterion only 16 Con. Banks and 6 Isl. Banks are selected. Data is extracted from the financial statements published at website of the State Bank of Pakistan and the banks’ websites. All the amounts used for inputs and outputs are in Pakistani rupees.
Data Envelopment Analysis