A comparison of chance-constrained DEA and stochastic frontier analysis: bank efficiency in Taiwan

  • PDF / 221,523 Bytes
  • 9 Pages / 595 x 794 pts Page_size
  • 5 Downloads / 178 Views

DOWNLOAD

REPORT


#2002 Operational Research Society Ltd. All rights reserved. 0160-5682/02 $15.00 www.palgrave-journals.com/jors

A comparison of chance-constrained DEA and stochastic frontier analysis: bank efficiency in Taiwan Tser-yieth Chen* Ming-chuan University, Taipei, Taiwan We employed both chance-constrained data envelopment analysis (CCDEA) and stochastic frontier analysis (SFA) to measure the technical efficiency of 39 banks in Taiwan. Estimated results show that there are significant differences in efficiency scores between chance-constrained DEA and stochastic frontier production function. The advanced setting of the chance-constrained mechanism of DEA does not change the instinctive differences between DEA and SFA approaches. We further find that the ownership variable is still a significant variable to explain the technical efficiency in Taiwan, irrespective of whether a DEA, CCDEA or SFA approach is used. Journal of the Operational Research Society (2002) 53, 492–500. DOI: 10.1057=palgrave=jors=2601318 Keywords: chance-constrained programming; stochastic frontier analysis; data envelopment analysis

Introduction In this paper, we compare chance-constrained data envelopment analysis (CCDEA) with stochastic frontier analysis (SFA) to estimate technical efficiency indices, and to explore the effect of ownership on the technical efficiency for policy purposes in the banking sector. We focus on the efficiency assessment since we believe that efficiency and=or performance will become strategic variables in tackling the increasing competitive pressure and structural changes within this industry. We incorporate a stochastic mechanism and employ both CCDEA and SFA approaches to our analysis since we consider that the unpredictability of customer demand makes the banks’ input–output relationship stochastic. There are two reasons: firstly, bank service often consists of one-off activities and the results of bank service processes are commonly of a stochastic nature. Both the nature of deposit=loans risk and the attitude toward deposit=loan risk are often different in different customers. Thus, the outputs of bank services provide various indicators of performance and they may be the result of inexperience, added to the intrinsic difficulty of prediction and the unpredictability of customer demand. Secondly, the turbulent Asian financial crisis period made market conditions more unpredictable and stochastic. The CCDEA and SFA results can suggest the presence of theoretical output variability components in efficiency levels, meaning that the best practice efficiency indices of each bank are relatively stable. We also examine the relationship between technical effi*Correspondence: Tser-yieth Chen, Institute of Management Science, Ming-chuan University, No 250, Chung-shan North Road, Section 5, Taipei, 11120, Taiwan. E-mail: [email protected]

ciency and bank ownership in order to provide some recommendations for bank management. Deregulation in the financial market is now perceived as a worldwide phenomenon and Taiwan is no excepti