Cost efficiency of top Thai banks: A comparison of classical stochastic frontier with efficiency stochastic frontier models

Abstract

© 2017 by the Mathematical Association of Thailand. All rights reserved. Without doubts, bank performance’s evaluation has increased its importance to bank managers and policymakers as a supportive factor that will help Thailand move forward. This paper conducts an empirical investigation into the consistency of efficiency scores given by the classical stochastic frontier and efficiency stochastic frontier models. We make an estimation of cost efficiency which is based on top four Thai banks data during the period 2001 to 2016. The results suggest that all of the coefficients from efficiency SFM fall between the 5 percent and 95 percent interval and are significant compared with classical SFM which have only some significant coefficients with overestimated and underestimated signs. Relatively, the efficiency stochastic frontier model outperforms the classical stochastic frontier model in terms of higher R-squared values. Based on our finding, it shows that KBANK’s TE is the only one that has moved closer to the frontier. Meanwhile, SCB bank’s TE moved sightly away from the frontier. The other banks; BBL and BAY banks tend to remain steady throughout the studied period. These characteristics can be used as supplementary information for bank managers in order to achieve their higher efficiency along with observing financial ratios concurrently. We strongly believe that efficiency stochastic frontier models is more efficient in terms of evaluations of bank performance.