250 likes | 399 Views
Has the banking sector become more vulnerable over time? Deniz Anginer Financial Economist, DECFP Asli Demirguc-Kunt Chief Economist FPD, Research Manager DECFP May 2011 Dubrovnik Economic Conference. Globalization.
E N D
Has the banking sector become more vulnerable over time? Deniz Anginer Financial Economist, DECFP Asli Demirguc-Kunt Chief Economist FPD, Research Manager DECFP May 2011 Dubrovnik Economic Conference
Globalization • The last two decades have seen a tremendous transformation in the global financial sector driven by: • Globalization • Innovations in financial engineering and communications technology • De-regulation • These global trends have resulted in: • Increased productivity • Increased capital flows • Lower cost of borrowing • Price discovery and more opportunities for risk diversification • Same trends have also resulted in increased concentration, complexity, and exposure to common sources of risk
What do we do? • We examine whether these global trends have lead to an increase in banks’ exposure to common risk factors • We construct a default risk measure for all publicly traded banks using the Merton (1974) contingent claim model • Create a weekly time series of default probabilities for over 2,000 banks in over 70 countries and examine the evolution of the covariance structure of default risk of banks over time • We also examine cross-country differences and link them various measures of financial and economic openness
What do we find? • Systematic default risk has a significant global component in the banking sector • There has been a significant increase in default risk co-dependence leading up to the 2007/2008 financial crisis • The increase in co-dependence has been higher for North American and European banks, as well as banks that are larger • Developing countries which are more integrated, have liberalized financial systems and weak banking supervision have higher co-dependence in their banking sector
Policy Implications • In the aftermath of the crisis of 2007/08, there has been renewed interest in macro-prudential regulation • There has also been a growing consensus to adjust capital requirements to better reflect an individual bank’s contribution to the risk of the financial system as a whole • Acharya et al (2010) Brunnermeier, Crockett, Goodhart, Persaud, and Shin (2009), Financial Stability Forum (2009) • Our results support an increase in scope for intra-national supervisory co-operation, as well as capital charges for too-connected-to-fail’ institutions that can impose significant externalities
Literature • Growing number of papers examine the risk of individual banks to the banking system • Acharya et al (2010), Adrian and Brunnermeier (2009), Huang et al. (2009), Chan-Lau and Gravelle (2005), Avesani et al. (2006), and Elsinger and Lehar (2008) • Others have examined the correlation structure of equity returns of a subsample of banks • De Nicolo and Kwast (2002), Schuler (2002), Hawkesby, Marsh and Stevens (2003,2007) • Larger contagion/convergence literature • Forbes and Rigobon (2002), Kee-Hong Bae and Stulz (2003), Bekeart and Wang (2009); Pukthuanthong and Roll (2009)
Average PD = 4% Default C - 4 - 3 - 2 - 1 0 1 2 3 4 Merton Model • We compute default probabilities implied from the structural credit risk model of Merton (1974) • Commonly used in default prediction outperforming accounting-based models in hazard regressions • Campbell, Hilscher and Szilagyi 2008; Hillegeist, Keating, Cram, and Lundstedt, 2004; Bharath and Shumway, 2008) • Merton (1977) points out the applicability of the contingent claims approach to pricing deposit insurance in the banking context. • Bongini, Laeven, and Majnoni (2002), Bartram, Brown and Hundt (2008) and others have used the Merton model to measure default probabilities of commercial banks Log Asset Value Distribution
Data Coverage • Market cap and equity volatility obtained from Datastream; Bank assets and liability information comes from BankScope • Our results are robust to alternative Distance-to-Default definitions • We impose a number of filters to ensure data integrity • Final sample includes 2029 banks from 70 countries starting in 1998
Global Component of Changes in Credit Risk • Principal components of log changes in Default probabilities for commercial banks Jan 1998 – Oct 2010 • The first component explains 60% of variation in changes in default risk of commercial banks
Clustering in Default Risk • This chart shows the percentage of banks in a given week that have simultaneous worst change in default risk over a 12 month time period
Decomposing the Systematic Changes in Default Risk • We follow Heston & Rouwenhorst (1994)’s method to decompose the systematic variance of changes in default risk into global and regional effects:
Increase in Bank Concentration • Concentration measures assets of 3 largest banks as a share of assets of all commercial banks • There has been a substantial increase in concentration in both developing and developed countries
Decomposing the Systematic Changes in Default Risk • Banks Size explains a significant portion of systematic variation in default risk • Co-dependence is significantly higher for larger banks
Variance Ratio • Variance ratio (Bekaert and Wang (2009), Ferreira and Gama (2005)): • Variance ratio calculated for all banks in the data set Jan 1998 – Oct 2010 period • Starting in 2004 there has been an upward trend leading up to the crises
Comovement • Co-movement (Harmon et al (2010): • Chart shows the distribution in a given year of the % of banks that had a positive increase in default probability
Quintile Regression • Quintile regression (Boyson, Stahel and Stulz (2010), Brunnermeier and Pedersen (2009): • Co-dependence is higher for higher levels of default risk changes
Trends in Co-dependence • We formally test to see if there has been a change in co-dependence over time • During the crisis there has been an increase in co-dependence for all banks in all regions • On average we find that starting in 2004 leading up to the crisis, there has been an upward increase in co-dependence • But there is much cross-country variation, which we explore next
Conclusion • We create a database of default risk measure for all publicly traded banks using the Merton (1974) contingent claim model • We show that systematic default risk has a significant global component in the banking sector and that there has been a significant increase in default risk co-dependence leading up to the 2007/2008 financial crisis • There is much cross-sectional variation, and countries which are more integrated, have liberalized financial systems and weak banking supervision have higher co-dependence in their banking sector • Our results support an increase in scope for intra-national supervisory co-operation, as well as capital charges for too-connected-to-fail’ institutions that can impose significant externalities