Efficiency Analysis of Commercial Banks in Pakistan – A Non Parametric Approach

72 Pages Posted: 16 Nov 2011 Last revised: 17 Nov 2011

See all articles by Saba Hayat

Saba Hayat

affiliation not provided to SSRN

Date Written: November 15, 2011


This study investigates the technical and profit efficiency of commercial banks in Pakistan. Data Envelopment Analysis is used as a preferred methodology to accurately determine profit and technical efficiency. The constant return to scale model is employed and Excel Solver is used to give scores which highlights the efficiency differences across three ownership categories. The study identifies loans and interest income as outputs and deposits and interest expense as the inputs to measure the technical efficiency. It identifies loan loss provision as a factor causing deviation in profit efficiency from the technical efficiency. Our study suggests that the profit efficiency improves over the years and technical efficiency is also rising in 2009. Large banks are relatively more efficient than small banks and private banks have better efficiency scores. The results can be generalized for commercial banks across different ownership structures and size but this generalization doesn’t spread across the Islamic banking sector due to the choice of DMUs which include no Islamic bank. According to the policy maker’s perspective, the study establishes the economic importance of banking industry in the country.

Keywords: Data Envelopment Analysis, Profit efficiency, Ownership, technical Efficiency, Regulatory reforms, Bank efficiency

JEL Classification: C00, C67, G21

Suggested Citation

Hayat, Saba Javed, Efficiency Analysis of Commercial Banks in Pakistan – A Non Parametric Approach (November 15, 2011). Available at SSRN: https://ssrn.com/abstract=1960063 or http://dx.doi.org/10.2139/ssrn.1960063

Saba Javed Hayat (Contact Author)

affiliation not provided to SSRN ( email )

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics