Two-stage DEA-Truncated Regression: Application in banking efficiency and financial development

Fernandes, F. D. S., Stasinakis, C. and Bardarova, V. (2018) Two-stage DEA-Truncated Regression: Application in banking efficiency and financial development. Expert Systems with Applications, 96, pp. 284-301. (doi: 10.1016/j.eswa.2017.12.010)

156248.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.



This study evaluates the efficiency of peripheral European domestic banks and examines the effects of bank-risk determinants on their performance over 2007–2014. Data Envelopment Analysis is utilised on a Malmquist Productivity Index in order to calculate the bank efficiency scores. Next, a Double Bootstrapped Truncated Regression is applied to obtain bias-corrected scores and examine whether changes in the financial conditions affect differently banks’ efficiency levels. The analysis accounts for the sovereign debt crisis period and for different levels of financial development in the countries under study. Such an application in the respective European banking setting is unique. The proposed method also copes with common misspecification problems observed in regression models based on efficiency scores. The results have important policy implications for the Euro area, as they indicate the existence of a periphery efficiency meta-frontier. Liquidity and credit risk are found to negatively affect banks productivity, whereas capital and profit risk have a positive impact on their performance. The crisis period is found to augment these effects, while bank-risk variables affect more banks' efficiency when lower levels of financial development are observed.

Item Type:Articles
Glasgow Author(s) Enlighten ID:Stasinakis, Dr Charalampos
Authors: Fernandes, F. D. S., Stasinakis, C., and Bardarova, V.
College/School:College of Social Sciences > Adam Smith Business School > Accounting and Finance
Journal Name:Expert Systems with Applications
ISSN (Online):1873-6793
Published Online:08 December 2017
Copyright Holders:Copyright © 2017 Elsevier Ltd.
First Published:First published in Expert Systems with Applications 96:284-301
Publisher Policy:Reproduced in accordance with the copyright policy of the publisher

University Staff: Request a correction | Enlighten Editors: Update this record