Bank efficiency, productivity, and convergence in EU countries: a weighted Russell directional distance model

Aarti Rughoo, Hidemichi Fujii, Shunshuke Managi, Roman Matousek

Research output: Contribution to journalArticlepeer-review

13 Citations (Scopus)
39 Downloads (Pure)

Abstract

The objective of this study is three-fold. First we estimate and analyse bank efficiency and productivity changes in the EU28 countries with the application of a novel approach, a weighted Russell directional distance model. Second, we take a disaggregated approach and analyse the contribution of the individual bank inputs on bank efficiency and productivity growth. Third, we test for convergence in EU28 bank productivity as well as in the inefficiency of individual bank inputs. We find that bank efficiency has been undermined by the financial crisis in banks notably from the EU15 countries. We also argue that bank efficiency and productivity in EU countries vary across the banking sector with banks from the ‘old’ EU showing higher efficiency levels. Nonetheless, a noticeable catching up process is observed for banks from the ‘new’ EU countries. Consequently, we do not find evidence of group convergence for bank productivity but there is evidence of convergence in bank efficiency change and technical change among the EU28 countries throughout the period 2005–2014. The driving force seems to be convergent technical change from the old EU member states’ banks. On the other hand, almost no convergence is detected for the banks’ individual inputs while the transition paths show heightened diversity during the crisis years.
Original languageEnglish
Pages (from-to)135-156
Number of pages22
JournalThe European Journal of Finance
Volume24
Issue number2
Early online date21 Mar 2017
DOIs
Publication statusPublished - 22 Jan 2018

Keywords

  • European Union
  • Bank efficiency
  • Convergence

Fingerprint

Dive into the research topics of 'Bank efficiency, productivity, and convergence in EU countries: a weighted Russell directional distance model'. Together they form a unique fingerprint.

Cite this