Elsevier

Journal of Banking & Finance

Volume 75, February 2017, Pages 184-199
Journal of Banking & Finance

Bank productivity growth and convergence in the European Union during the financial crisis

https://doi.org/10.1016/j.jbankfin.2016.11.016Get rights and content

Highlights

  • Bank productivity growth and integration process for the 28 EU countries during the financial crisis.

  • Innovative methodological framework based on the Malmquist Productivity Index extended by an additive two-stage DEA model.

  • β-convergence and σ-convergence of the two-stage Productivity Index.

  • Loss of competitiveness of the European banking system due to the drop in growth of the performance stage and technical change.

  • Strong convergence pattern during the financial crisis.

Abstract

This paper examines the bank productivity growth and integration process for the 28 EU countries during three main phases of the financial crisis: the U.S. subprime crisis (2007–2008), the global financial crisis (2009–2010) and the sovereign debt crisis (2010–2012). We extend the Malmquist Productivity Index by applying an additive two-stage DEA model. This allows us to explore the sources of growth in different stages of production. Furthermore, we assess the integration of European banks by analyzing the β-convergence and σ-convergence of the two-stage Productivity Index. Our results show a productivity growth during the U.S. subprime crisis, but a consistent decline during the global financial crisis. The loss of competitiveness of the European banking system is due to the drop in growth of the performance stage and technical change. Finally, we find a strong convergence pattern during the financial crisis, mainly driven by the catch up process of some Eastern countries and the drop in performance of Western countries.

Introduction

The recent Global Financial Crisis (GFC) and Eurozone sovereign debt crisis have severely undermined bank stability and destabilized the process of bank integration in Europe. From the onset of the financial crisis, national public authorities have adopted wide-ranging interventions, such as recapitalization, providing debt guarantees, and conducting asset purchases, to reduce the fragility of the banking system and restore confidence in the financial markets. Even though these interventions could have been beneficial for the stability of the banking system, they have also raised concerns as regards their impact on bank competition and consequently banking integration (e.g. Berger and Bouwman, 2013, Calderon and Schaeck, 2015). The global financial crisis (GFC) appears in fact to have contributed to a leveraging of the existing differences and fragmentation between countries in Europe, and slowed down the banking integration process (Matousek et al., 2015). More generally, the GFC has shown the fragility of the monetary union and the lack of a sufficient degree of financial and fiscal integration to react promptly to negative macroeconomic shocks (Lane, 2012). During the years 2007–2012, the European banking system went through three main financial crisis’ phases that have exerted a heterogeneous and asymmetric impact across countries and especially the Euro area (Hristov et al., 2012). For example, countries with a great reliance on external funds, and especially on short-term debt markets, such as Ireland, were strongly affected by the global financial crisis (Milesi-Ferretti and Tille, 2011).

According to Aı¨t-Sahalia et al., (2012) and recently by Fiordelisi and Ricci (2015), the financial crisis period can be divided in the following three phases: the subprime financial crisis (2007–2008), the global financial crisis (2008–2010) and the sovereign debt crisis (2010–2012). At the beginning of the subprime financial crisis in 2007, the central banks’ interventions seemed to work effectively (Fiordelisi and Ricci, 2015). Despite the large losses in the subprime mortgage market, the overall impression was that the worst was going to end soon (see Mishkin, 2011). However, in late 2008, as a consequence of the collapse of large US financial institutions such as Lehman Brothers and AIG and the run on the Reserve Primary Fund, the financial crisis worsened (Mishkin, 2011). In this phase, the financial crisis started to affect Europe as much as the United States and it became “global”. Through 2008 and 2009, the focus was on the interventions put forward by the European Central Bank in response to the global financial shock rather than on the country-specific financial risks (Lane, 2012). At that time there were no indications of a profound European sovereign debt crisis that since the second half of 2009 plunged economies in some EU countries into a deep recession, e.g. Greece, Ireland, Italy, Portugal, and Spain (GIIPS). These countries experienced a raise in deficit/GDP ratio and increasing estimates of prospective banking sector losses on bad loans (Mody and Sandri, 2012). The most shocking news was however originated in Greece with the revelation of the extreme violation of the euro's fiscal rules (Broner et al., 2014, Lane, 2012). All these events contribute to a sharp deterioration of the situation and a raising of the spreads on sovereign bonds between Germany and GIIPS countries. Nonetheless, not only the GIIPS countries suffered as a result of the sovereign debt crisis, but the entire Eurozone was affected by this phase of the crisis. The government bond markets in the Eurozone became more fragile and more susceptible to self-fulfilling liquidity crises, as they were associated with negative sentiments that were strong at the end of 2010 (De Grauwe and Ji, 2013). Even the European countries with the best competitive and growth prospects experienced an increase in spreads in a significant manner when their financial sectors were under stress (Mody and Sandri, 2012). However, such an effect was higher for countries facing a large debt burden.

All these adverse developments and events have impacted on the banks’ productivity and harmed the process of integration as some countries were more vulnerable than others. The financial crisis exerted a pervasive pressure on banking system and undermined the banking activities from both the funding side and the lending side. From the funding side, banks were under pressure because of the freezing of the European Interbank market, and the threat of the drop of the deposit supply side because of “bank runs” (Iyer et al., 2014). From the lending supply, banks appear to have consistently reduced their lending activities and new loans (e.g. De Haas and Van Horen, 2013, Ivashina et al., 2013). In addition, over the period 2008–2012, the EU-27 banking and financial received state aids in terms of total recapitalization and asset relief for EUR 1337.26 bn, and EUR 3931.71bn in terms of total guarantees and liquidity measures.1

In this context, some important empirical questions arise: How did the productivity of European banks change during the three phases of the crisis: The U.S. subprime crisis (2007–2008), the global financial crisis (2009–2010) and the sovereign debt crisis (2010–2012)? What were the sources of productivity change in Europe during the crisis? How and to what extent did the financial crisis affect the convergence process of the European banking system? How did the GIIPS countries and the countries in the Euro zone behave compared to the banks in other EU countries?

This paper address these issues and provides empirical evidence on these unexplored avenues of research. It specifically contributes to the existing literature on productivity and convergence in three ways. Firstly, we examine the patters of productivity growth. While previous papers mainly focus on the link between efficiency and integration (Weill, 2009, Casu and Girardone, 2010, Matousek et al., 2015), they divert their attention away from the sources of productivity growth. The productivity index encompasses different sources of growth that can play a pivotal role for the economic recovery of Europe after the financial crisis. Not only bank efficiency, but innovation is a relevant driver for economic growth as well (Aghion and Howitt, 1998, Grossman and Helpman, 1991). Therefore, our aim is to assess the sources of productivity growth and the underlying processes of European banking integration during the financial crisis.

Secondly, a further contribution of our study lies in providing new evidence on the drivers underlying the bank productivity and converge patterns between “Old Europe” and “New Europe” during the crisis period. The recent financial crisis could have slowed down the integration process of the “New Europe” with the “Old Europe” . Transaction economies have recently suffered from a sharp drop in the rate of investments. Moreover, there was a consistent deleveraging process in the banking system, which has contributed to a widening of the credit crunch of industrial firms (EBRD, 2015). In addition, the turmoil in the Euro zone exerted a negative effect on the growth of transition regions, especially for those countries whose banking sector is deeply integrated with Eurozone-based banks.2

Thirdly, this paper introduces an innovative methodological approach to examine the productivity and convergence of the EU banking system. To our best knowledge, this is the first paper that examines the productivity and convergence of the EU banking system by applying a two-stage data envelopment analysis (DEA) model. We apply a relational two-stage DEA model in order to assess the components of the productivity index. This model allows us to overcome the classical deposit dilemma that affects the identification strategy of empirical studies on bank efficiency. It in fact treats deposits as an intermediate variable (Fukuyama and Matousek, 2011, Fukuyama and Weber, 2010, Holod and Lewis, 2011). In other words, we use deposits as output in a first stage where employees and capital are inputs, and as input to produce financial assets (loans and other earning assets) in a second stage. In particular, we employ the Malmquist productivity index to evaluate the productivity of 539 commercial banks in the EU-28 countries during three periods of the financial crisis. We first disaggregate the productivity into efficiency change, technical change, and scale change using Ray and Desli's (1997) decomposition. Furthermore, drawing on Kumar and Russell (2002) we investigate the convergence of the overall bank productivity and the productivity of the above two stages. By disentangling the production process of a bank, we can better identify different sources of transmission of inefficiency or decline of innovation in the European banking system. This new approach enable us to examine both the sources of productivity for both banks’ lending and funding activities. This is important to fully address our research questions.

The structure of the paper is as follows. Section 2 reviews the recent literature about productivity and convergence in the first part and two-stage DEA models and deposits dilemma in the second part. Section 3 describes the methodology; Section 4 presents and discusses the empirical results Last, Section 5 concludes.

Section snippets

Literature review

This section reviews the recent literature on the convergence of European banks and serves as a theoretical background for the examination of total factor productivity in banking industry and the two-stage DEA models and their application to banks.

The additive two-stage DEA model

In this section, we outline our methodological approach. The proposed model is based on the additive efficiency decomposition approach proposed by Chen et al., (2009).3 Given n DMUs, xij(i=1,,m), zdj(d=1,,D) and yrj(r=1,,s) are respectively the ith input, the dth intermediate variable, and the rth output respectively of the jth DMU (j=1,,n). Moreover, vi, wd and yr are the multipliers of the model. The overall efficiency E0 is in the

Data and model description

We collected unconsolidated annual income statements and balance sheet data from the Fitch-IBCA BankScope (BSC) database. For our analysis, we only include commercial banks from 28 EU countries. We dropped banks with negative equity values and missing values for total assets. Hence, our sample consists of 539 commercial banks for 28 EU countries. All data is deflated to 2010 prices. Following Fukuyama and Weber (2010), we construct a two-stage banking efficiency model in order to keep the dual

Conclusion

In this paper, we investigate the sources of bank productivity growth and the underlying patters of the integration process for the 28 EU countries during the financial crisis. By combining the Malmquist Productivity Index and an additive two-stage DEA model we explore the sources of growth in different stages of production and we quantify the contribution of individual components (inputs and outputs) to productivity change. We then investigate the convergence of the overall bank productivity

Acknowledgement

We would like to thank the two anonymous reviewers for their constructive comments on an earlier version of our paper. Any remaining errors are solely the authors’ responsibility.

References (91)

  • F. Broner et al.

    Sovereign debt markets in turbulent times: creditor discrimination and crowding-out effects

    J. Monet. Econ.

    (2014)
  • B. Casu et al.

    Integration and efficiency convergence in EU banking markets

    Omega

    (2010)
  • B. Casu et al.

    Productivity change in european banking: a comparison of parametric and non-parametric approaches

    J. Bank. Financ.

    (2004)
  • T.-P. Chang et al.

    The sources of bank productivity growth in china during 2002-2009: a disaggregation view

    J. Bank. Financ.

    (2012)
  • A Charnes et al.

    Measuring the efficiency of decision making units

    Eur. J. Oper. Res.

    (1978)
  • Y. Chen et al.

    Additive efficiency decomposition in two-stage DEA

    Eur. J. Oper. Res.

    (2009)
  • P. De Grauwe et al.

    Self-Fulfilling crises in the Eurozone: an empirical test

    J Int Money Finance

    (2013)
  • M.D. Fethi et al.

    Assessing bank efficiency and performance with operational research and artificial intelligence techniques: a survey

    Eur. J. Oper. Res.

    (2010)
  • F. Fiordelisi et al.

    The effect of monetary policy interventions on interbank markets, equity indices and G-SIFIs during the crisis

    J. Financ. Stab.

    (2014)
  • H. Fukuyama et al.

    Efficiency of Turkish banking: two-stage network system. Variable returns to scale model

    J. Int. Financ. Markets Institut. Money

    (2011)
  • H. Fukuyama et al.

    A slacks-based inefficiency measure for a two-stage system with bad outputs

    Omega

    (2010)
  • J. Goddard et al.

    European banking: an overview

    J. Bank. Finance

    (2007)
  • E. Grifell-Tatje et al.

    The sources of productivity change in Spanish banking

    Eur. J. Oper. Res.

    (1997)
  • G.E. Halkos et al.

    A unified classification of two-stage DEA models

    Surv. Oper. Res. Manag. Sci.

    (2014)
  • D. Holod et al.

    Resolving the deposit dilemma: a new DEA bank efficiency model

    J. Bank. Financ.

    (2011)
  • N. Hristov et al.

    Loan supply shocks during the financial crisis: evidence for the Euro area

    J. Int. Money Finance

    (2012)
  • I. Isik et al.

    Financial deregulation and total factor productivity change: an empirical study of Turkish commercial banks

    J. Bank. Financ

    (2003)
  • C. Kao et al.

    Efficiency decomposition in two-stage data envelopment analysis: an application to non-life insurance in Taiwan

    Eur. J. Oper. Res.

    (2008)
  • C. Kao et al.

    Multi-period efficiency and malmquist productivity index in two-stage production systems

    Eur. J. Oper. Res.

    (2014)
  • A. Koutsomanoli-Filippaki et al.

    Efficiency and productivity growth in the banking industry of central and eastern Europe

    J. Bank. Financ

    (2009)
  • R. Matousek et al.

    Bank performance and convergence during the financial crisis: evidence from the “old” European union and Eurozone

    J. Bank. Financ

    (2015)
  • A. Rughoo et al.

    Integration in European retail banking: evidence from savings and lending rates to non-financial corporations

    J. Int. Financ. Markets Institut. Money

    (2012)
  • X. Sala-i-Martin

    Regional cohesion: evidence and theories of regional growth and convergence

    Eur. Econ. Rev.

    (1996)
  • J.-E. Sturm et al.

    Foreign bank entry, deregulation and bank efficiency: lessons from australian experience

    J. Bank. Financ.

    (2004)
  • N.G. Tzeremes

    Efficiency dynamics in indian banking: a conditional directional distance approach

    Eur. J. Oper. Res.

    (2015)
  • K. Wang et al.

    Efficiency measures of the Chinese commercial banking system using an additive two-stage DEA

    Omega

    (2014)
  • P. Wanke et al.

    Two-stage DEA: an application to major Brazilian banks

    Expert Syst. Appl.

    (2014)
  • P. Wanke et al.

    Financial distress drivers in Brazilian banks: a dynamic slacks approach

    Eur. J. Oper. Res.

    (2015)
  • L. Weill

    Convergence in banking efficiency across european countries

    J. Int. Financ. Markets Institut. Money

    (2009)
  • Y-C. Wu et al.

    The impact of earnings management on the performance of ASEAN banks

    Econ. Model.

    (2016)
  • P. Aghion et al.

    Endogeneous Growth Theory

    (1998)
  • I.M.S. Alam

    A non-parametric approach for assessing productivity dynamics of large banks

    J. Money Credit Bank.

    (2001)
  • Y. Altunbas et al.

    Technical change in banking

    Econ. Lett.

    (2001)
  • M.G. Arghyrou

    The global financial crisis in Greece: its background causes, escalation and prospects for recovery

  • A.G. Assaf et al.

    Productivity and efficiency analysis of shinkin banks: evidence from bootstrap and Bayesian approaches

    J. Bank. Financ.

    (2011)
  • Cited by (0)

    View full text