Federal Deposit InsuranceCorporation• Center for Financial Researchh
Sanjiv R. Das
Darrell Duffie
Nikunj Kapadia
Risk-Based Capital Standards,
Deposit Insurance and Procyclicality
Risk-Based Capital Standards,
Deposit Insurance and Procyclicality
FDIC Center for Financial Research
Working Paper
No. 2005-12
Estimating Systemic Risk in the International Financial System
Söhnke M. Bartram
Gregory W. Brown
John E. Hund
July 2005
Sanjiv R. Das
Darrell Duffie
Nikunj Kapadia
Risk-Based Capital Standards,
Deposit Insurance and Procyclicality
Risk-Based Capital Standards,
Deposit Insurance and Procyclicality
FDIC Center for Financial Research
Working Paper
No. 2005-12
Estimating Systemic Risk in the International Financial System
Söhnke M. Bartram
Gregory W. Brown
John E. Hund
July 2005
Estimating Systemic Risk in the International Financial System
Söhnke M. Bartram*, Gregory W. Brown†, and John E. Hund‡
FDIC Center for Financial Research Working Paper No. 2005-12
Abstract
Using a unique and comprehensive dataset, this paper develops and uses three distinct methods to
quantify the risk of a systemic failure in the gl obal banking system. We examine a sample of 334
banks (representing 80% of global bank equity) in 28 countries around 6 global financial crises (such
as the Asian and Russian crises and September 11, 2 001), and show that these crises did not create
large probabilities of global financial system failure. First, we show that cumulative negative abnor-
mal returns for the subset of banks not directly ex posed to a negative shock (unexposed banks)
rarely exceed a few percent. Second, we use structural models to obtain more precise point estimates
of the likelihood of systemic failure. These estimates suggest that systemic risk is limited even during
major financial crises. For example, maximum like lihood estimation of bank failure probabilities im-
plied by equity prices suggests the Asian crisis induced less than a 1% increase in the probability of
systemic failure. Third, we also obtain estimates of systemic risk implied by equity option prices of
U.S. and European banks. The largest values are obtained for the Russian crisis and September 11
and these show increases in estimated average de fault probabilities of only around 1-2%. Taken to-
gether our results suggest statistically significant, but economically small, increases in systemic risk
around even the worst financial crises of the la st 10 years. Although policy responses are endoge-
nous, the low estimated probabilities suggest that the distress of central bankers, regulators and poli-
ticians about the events we study may be overstated, and that current policy responses to financial
crises and the existing institutional framework may be adequate to handle major macroeconomic
events.
July 15, 2005
Keywords: Systemic risk, default risk, credit risk, banks, exposure, emerging markets, 9/11
JEL Classification: G3, F4, F3
CFR research programs: Deposit Insurance
* Lancaster University, Management School, Lancaster LA1 4YX, United Kingdom, Phone: +44 (1524) 592 083, Fax: +1 (425)
952 10 70, Email: <s.m.bartram@lancaster.ac.uk>, Internet: <http://www.lancs.ac.uk/staff/bartras1/>
† Corresponding author. Kenan-Flagler Business School, University of North Carolina at Chapel Hill, CB 3490 McColl Build-
ing, Chapel Hill, NC 27599-3490, USA, Phone: +1 (919) 962-9250, Fax: +1 (919) 962-2068, Email:
<gregwbrown@unc.edu>, Internet: <http://intranet.kenan-flagler.unc.edu/faculty/browngr/>.
‡ McCombs School of Business, University of Texas at Austin, Austin, TX 78712, USA. Phone: +1 (512) 471-6587, Fax: +1
(512) 471-5073, Email: <jhund@mail.utexas.edu>.
The authors gratefully acknowledge research funding by the Center for Financial Research of the FDIC, Lancaster Univer-
sity, Lancaster University Management School, the Leverhulme Trust, and The Bank of America Faculty Development Fund,
as well as support by Global Reports, Perfect Information, Fitch Ratings and Thomson Financial in establishing the dataset.
The authors also thank Robert Bliss, Jean Dermine, Eva Gu tierrez, Philipp Hartmann, Shelag h Heffernan, Jean Helwege,
Armen Hovakimian, Donald Morgan, Robe rto Rigobon, Lynn Shibut, Christian Uppe r and Giuseppe Vulpes as well as
seminar participants at the FDIC Summer Research Workshop, the FIRS conferen ce in Capri, the 2005 American Finance
Association Annual Meeting, the 2004 European Finance Association Annual Meeting, the 2004 Financial Management As-
sociation Annual Meeting, the EMG/ESRC workshop on "Inter national Diversification and Contagion" at Cass Business
School, The Bank of England, Darden School of Business, La ncaster University, Temple Univ ersity, University of North
Carolina, University of Texas at Austin, Vanderbilt University and Wake Forest University. Florian Bardong and Nishad Ka-
padia provided excellent research assistance.
Söhnke M. Bartram*, Gregory W. Brown†, and John E. Hund‡
FDIC Center for Financial Research Working Paper No. 2005-12
Abstract
Using a unique and comprehensive dataset, this paper develops and uses three distinct methods to
quantify the risk of a systemic failure in the gl obal banking system. We examine a sample of 334
banks (representing 80% of global bank equity) in 28 countries around 6 global financial crises (such
as the Asian and Russian crises and September 11, 2 001), and show that these crises did not create
large probabilities of global financial system failure. First, we show that cumulative negative abnor-
mal returns for the subset of banks not directly ex posed to a negative shock (unexposed banks)
rarely exceed a few percent. Second, we use structural models to obtain more precise point estimates
of the likelihood of systemic failure. These estimates suggest that systemic risk is limited even during
major financial crises. For example, maximum like lihood estimation of bank failure probabilities im-
plied by equity prices suggests the Asian crisis induced less than a 1% increase in the probability of
systemic failure. Third, we also obtain estimates of systemic risk implied by equity option prices of
U.S. and European banks. The largest values are obtained for the Russian crisis and September 11
and these show increases in estimated average de fault probabilities of only around 1-2%. Taken to-
gether our results suggest statistically significant, but economically small, increases in systemic risk
around even the worst financial crises of the la st 10 years. Although policy responses are endoge-
nous, the low estimated probabilities suggest that the distress of central bankers, regulators and poli-
ticians about the events we study may be overstated, and that current policy responses to financial
crises and the existing institutional framework may be adequate to handle major macroeconomic
events.
July 15, 2005
Keywords: Systemic risk, default risk, credit risk, banks, exposure, emerging markets, 9/11
JEL Classification: G3, F4, F3
CFR research programs: Deposit Insurance
* Lancaster University, Management School, Lancaster LA1 4YX, United Kingdom, Phone: +44 (1524) 592 083, Fax: +1 (425)
952 10 70, Email: <s.m.bartram@lancaster.ac.uk>, Internet: <http://www.lancs.ac.uk/staff/bartras1/>
† Corresponding author. Kenan-Flagler Business School, University of North Carolina at Chapel Hill, CB 3490 McColl Build-
ing, Chapel Hill, NC 27599-3490, USA, Phone: +1 (919) 962-9250, Fax: +1 (919) 962-2068, Email:
<gregwbrown@unc.edu>, Internet: <http://intranet.kenan-flagler.unc.edu/faculty/browngr/>.
‡ McCombs School of Business, University of Texas at Austin, Austin, TX 78712, USA. Phone: +1 (512) 471-6587, Fax: +1
(512) 471-5073, Email: <jhund@mail.utexas.edu>.
The authors gratefully acknowledge research funding by the Center for Financial Research of the FDIC, Lancaster Univer-
sity, Lancaster University Management School, the Leverhulme Trust, and The Bank of America Faculty Development Fund,
as well as support by Global Reports, Perfect Information, Fitch Ratings and Thomson Financial in establishing the dataset.
The authors also thank Robert Bliss, Jean Dermine, Eva Gu tierrez, Philipp Hartmann, Shelag h Heffernan, Jean Helwege,
Armen Hovakimian, Donald Morgan, Robe rto Rigobon, Lynn Shibut, Christian Uppe r and Giuseppe Vulpes as well as
seminar participants at the FDIC Summer Research Workshop, the FIRS conferen ce in Capri, the 2005 American Finance
Association Annual Meeting, the 2004 European Finance Association Annual Meeting, the 2004 Financial Management As-
sociation Annual Meeting, the EMG/ESRC workshop on "Inter national Diversification and Contagion" at Cass Business
School, The Bank of England, Darden School of Business, La ncaster University, Temple Univ ersity, University of North
Carolina, University of Texas at Austin, Vanderbilt University and Wake Forest University. Florian Bardong and Nishad Ka-
padia provided excellent research assistance.