The efficiency of the banking system in Jordan, 1990-1996

Abstract

This study is an attempt to examine the efficiency of the banking system in Jordan during the period of 1990-1996. The study presents a comparative analysis of the frontier cost efficiency methodologies by applying econometric and mathematical programming techniques to a data set consisting of 20 banks over the period between 1990-1996. The study also analyzes the profit efficiency of the Jordanian banks during the same period of time by estimating a non-standard profit function. Furthermore, the study examines several possible sources of inefficiency in the Jordanian banks. For the panel data obtained during the period 1990-1996 for the Jordanian banks, Cobb-Douglas stochastic frontier was found to be an adequate representations of the data, given the specifications of the translog stochastic frontiers for banks in Jordan. The results indicated that the choice of efficiency estimation techniques makes a significant difference in terms of the cost efficiency scores. Efficiency scores obtained from the linear programming approach were found to be higher than the efficiency scores obtained from the econometric approach. The rankings were less well preserved between the econometric and the linear programming methodologies (rank correlation is 67.2%). Furthermore, the findings indicated that the average overall cost efficiency during the period of 1990-1996, which was 77.5% or 73.5% based on the econometric frontier approach and the mathematical programming approach respectively, can be improved substantially by improving both the technical and allocative efficiencies, which were 76.6% and 96.1%, respectively. Since the technical inefficiency dominates allocative inefficiency, cost inefficiency in the banking system in Jordan may, to a greater extent, be attributed to underutilization or wasting of inputs rather than choosing the incorrect input combinations. Technical inefficiency could be an indicator of the weak market forces induced by the market structure or the regulation that allows bank management to become remiss and continue their inefficient behavior. Foreign banks were found to be more efficient than national banks and small banking firms were more efficient than medium size and large banks. The scale inefficiency for large banks was found to be dominating the pure technical inefficiency, whereas, pure technical inefficiency was found to be dominating scale inefficiency for small and medium size banks. Furthermore, the study found that the overall estimated alternative profit inefficiency averaged 0.674, large banks are more profit efficient than other categories of banks, as they averaged 0.767, followed by medium size banks and then by small banks. The findings of the second stage regression indicated that banks with fewer employees per total assets, lower ratio of branch per total deposits, higher asset and higher salaries to total assets ratios were most cost efficient banks. The non-linear factors were found to be significant indicating the nonlinear effect of theses variable on the cost efficiency. The profit efficiency was found to be positively correlated with the age of the bank, size of the bank and the growth in bank total assets. However, the profit efficiency was found to be negatively correlated with the ratios of branch to total deposits, number of employees to total assets, salaries to total assets and risks. The research recommends some reforms in present policy towards the banking system. These changes include adopting a policy of encouraging foreign direct investment in the banking sector and eliminating all barriers facing this investment if any exist. Also, the study recommends that closer attention be given to credit risk by both bank management and the Central Bank of Jordan. Moreover, the study recommends that any banking reform in Jordan be aimed at increasing competitiveness in the banking system and achieving any increase in the capital of banks by means other than merger and consolidation, because mergers in a highly concentrated banking system may lead to fewer banks, which in turn will generally lead to less competitive performance.