Market Discipline under Systemic Risk Market Discipline under Systemic Risk Speaker: Sergio Schmukler Seventh Annual International Seminar on Policy Challenges for the Financial Sector Disclosure and Market Discipline May 30, 2007 World Bank, IMF, Federal Reserve Board
Background Work de la Torre, A., Levy-Yeyati, E., and Schmukler, S., 2003. Living and Dying with Hard Pegs: The Rise and Fall of Argentina s Currency Board, Economia, Spring, pp. 43-107. Levy-Yeyati, E., Martínez Pería, M. S., and Schmukler, S., 2004a. Market Discipline in Emerging Economies: Beyond Bank Fundamentals, in Market Discipline across Countries and Industries, C. Borio, W. Hunter, G. Kaufman, and K. Tsatsaronis (eds.), Cambridge: MIT Press. Levy-Yeyati, E., Martínez Pería, M. S., and Schmukler, S., 2004b. Market Discipline under Systemic Risk: Evidence from Bank Runs in Emerging Economies, World Bank Policy Research Working Paper No. 3440. Martínez Pería, M. S. and Schmukler, S., 2001. Do Depositors Punish Banks for Bad Behavior? Market Discipline, Deposit Insurance, and Banking Crises, Journal of Finance, 56:3, pp. 1029-1051. 1
Notion of Market Discipline Market discipline: economic agents face costs that increase with bank risk and react on the basis of these costs (Berger, 1991) Depositors respond to increases in bank risk via prices (interest rates) or quantities (deposits) This penalizes managers for excessive risk taking and disciplines them ex-ante Market discipline leads to a healthier banking system and a lower probability of banking crises 2
Crises in Emerging Economies Crises in emerging economies mostly related to prevalent systemic factors, rather than bank idiosyncratic risk The literature has mostly ignored how systemic factors influence market responses 3
Systemic Risk and Market Responses Systemic factors affect market responses, directly (regardless of bank-specific risk) Currency risk and deposit flight to a foreign currency Diminished deposit insurance and lender of last resort capabilities Dual agency problem Government interventions may affect the value of deposits through actions, beyond the control of bank managers 4
Systemic Risk and Market Responses Systemic factors affect market responses through bank idiosyncratic risk Exposure to country risk Public sector credit risk (bonds and loans) Exposure to currency risk (currency mismatch) Bank-level mismatch between assets and liabilities Debtor currency mismatch, private sector credit risk Bank liabilities mismatch Effect of systemic factors on future fundamentals 5
Our Work Study how systemic risk affects market responses through different channels Evidence from emerging market crises Systemic risk is essential to understand how depositors behave and extent of market discipline A re-examination of the concept of market discipline is necessary for emerging economies 6
Some Evidence 1. Two Banking Crises: A Brief Account 2. Role of Systemic Risk During Crises 3. Market Response to Bank Exposure to Systemic Risk 4. Systemic Risk and Bank Fundamentals 7
September 2000-December 2001 60,000 50,000 40,000 30,000 20,000 10,000 0 Sep-00 Nov-00 Jan-01 Mar-01 May-01 Jul-01 Sep-01 Nov-01 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0 8 Million U.S. Dollars Basis Points Argentina: Deposits and Systemic Factors System Time Deposits (left axis) Exchange Rate Risk (right axis) Country Risk (right axis)
Argentina: Change in Time Deposits by Bank 25% June 2001-December 2001 Peso Time Deposits 0% -25% -50% -75% -100% 25% Dollar Time Deposits 0% -25% -50% -75% -100% 9
Argentina: Change in Interest Rate by Bank 15% June 2001-December 2001 Peso Deposits 10% 5% 0% -5% 15% Dollar Deposits 10% 5% 0% -5% 10
December 2001-July 2002 12,000 10,000 8,000 6,000 4,000 2,000 0 Jan-01 Mar-01 May-01 Jul-01 Sep-01 Nov-01 Jan-02 Mar-02 May-02 Jul-02 5,000 4,500 4,000 3,500 3,000 2,500 2,000 1,500 1,000 500 0 11 Million U.S. Dollars Basis Points Uruguay: Deposits and Systemic Factors System Time Deposits (left axis) Exchange Rate Risk (right axis) Country Risk (right axis)
Uruguay: Change in Time Deposits by Bank December 2001 - July 2002 Dollar Time Deposits 25% 0% -25% -50% -75% -100% 12
7% 5% 3% 1% -1% -3% -5% Uruguay: Change in Interest Rate by Bank December 2001 - July 2002 Dollar Time Deposits 13
Some Evidence 1. Two Banking Crises: A Brief Account 2. Role of Systemic Risk During Crises 3. Market Response to Bank Exposure to Systemic Risk 4. Systemic Risk and Bank Fundamentals 14
Explanatory Power of Systemic Factors Percentage of Within Variance Explained by Country and Exchange Rate Risks Argentina Pre-Crisis Period Crisis Period Oct-98 - Jul-00 Sep-00 - Dec-01 Log of Peso Time Deposits 11.9% 70.9% Log of Dollar Time Deposits 2.4% 76.3% Uruguay Pre-Crisis Period Crisis Period Feb-99 - Dec-00 Jan-01 - Nov-01 Dec-01 - Jul-02 Log of Dollar Time Deposits 0.8% 5.2% 56.7% The percentage of within variance explained by country and exchange rate risks is calculated as one less the within R- squared of a fixed-effects regression including only bank fundamentals over the within R-squared of a fixed-effects regression including both bank fundamentals and country and exchange rate risks. In the case of Uruguay, both regressions also include total deposits in Argentina. 15
Argentina: Deposit Response Response to One Standard Deviation Shock in News Peso Time Deposits Dollar Time Deposits 0.5% 0.3% 0.1% -0.1% -0.3% -0.5% -0.7% -0.9% -1.1% -1.3% -1.5% 1 3 5 7 9 11 13 15 17 19 0.5% 0.3% 0.1% -0.1% -0.3% -0.5% -0.7% -0.9% -1.1% -1.3% -1.5% 1 3 5 7 9 11 13 15 17 19 Days Days Impulse response functions based on a VAR with ten lags. The model is estimated using daily data for 2000 and 2001. 16
Uruguay: Deposit Response Response to One Standard Deviation Shock in News Dollar Deposits 0.2% 0.0% -0.2% -0.4% -0.6% -0.8% -1.0% -1.2% 1 2 3 4 5 6 7 8 9 10 Days Impulse response functions based on a VAR with ten lags. The model is estimated using daily data from February 28, 2002 to July 29, 2002 17
Response to Crisis Shocks Cumulative Response of Deposits to the Five Largest Shocks in Each Series Argentina - 20-day Cumulative Response News Country Risk Exchange Rate Risk Combined Response System Peso Time Deposits -19.2% -11.5% -15.5% -46.1% System Dollar Time Deposits -9.7% -4.3% -5.6% -19.6% Uruguay - 10-day Cumulative Response News Country Risk Combined Response System Dollar Deposits -7.7% -7.0% -14.7% The combined response column shows the effects of joint shocks to news, country risk, and exchange rate risk. Combined responses are calculated by adding the percentage changes. 18
Some Evidence 1. Two Banking Crises: A Brief Account 2. Role of Systemic Risk During Crises 3. Market Response to Bank Exposure to Systemic Risk 4. Systemic Risk and Bank Fundamentals 19
Market Response to Bank Exposure Change in Time Deposits Average Percent Change in Time Deposits by Bank Exposure to Exchange Rate Risk (Dollar Liabilities/Total Liabilities) Argentina (September 2000 - December 2001) Uruguay (December 2001 - July 2002) -30.5-45.3-64.3-76.3-53.5 Peso Time Deposits Dollar Time Deposits Banks with Low Exposure Banks with High Exposure -81.2 Dollar Time Deposits 20
Some Evidence 1. Two Banking Crises: A Brief Account 2. Role of Systemic Risk During Crises 3. Market Response to Bank Exposure to Systemic Risk 4. Systemic Risk and Bank Fundamentals 21
0.9 0.8 0.6 0.5 0.3 0.2 0.0 Argentina: Systemic Risk and Bank Fundamentals Predictive Power of Lagged Fundamentals and Systemic Risk NPL and Country Risk NPL and Exchange Rate Risk 0.9 Country Risk (t-k) 0.8 0.6 NPL (t-k) NPL (t-k) 0.5 0.3 Exchange Rate Risk (t-k) 0.2 0.0 1 2 3 4 1 2 3 4 k k Point estimates of the coefficients from regressions of the non-performing loans ratio on its lagged values and systemic risk factors, estimated using Arellano and Bond's (1991) GMM procedure. 22
1.8 1.5 1.3 1.0 0.8 0.5 0.3 0.0 Uruguay: Systemic Risk and Bank Fundamentals Predictive Power of Lagged Fundamentals and Systemic Risk NPL and Country Risk NPL and Exchange Rate Risk 0.9 0.8 Country Risk (t-k) 0.6 NPL (t-k) NPL (t-k) 0.5 0.3 Exchange Rate Risk (t-k) 0.2 0.0 1 2 3 4 1 2 3 4 k k Point estimates of the coefficients from regressions of the non-performing loans ratio on its lagged values and systemic risk factors, estimated using Arellano and Bond's (1991) GMM procedure. 23
Conclusions Systemic factors not relevant during tranquil times, but are main drivers of market responses (deposit behavior) during crisis periods Even after controlling for bank fundamentals Using low and high frequency data A few large systemic shocks can destabilize the financial sector 24
Conclusions Deposit withdrawals during crises associated with bank exposure to exchange rate risk Its relevance increases as the crisis intensifies Exposure to systemic factors not priced Systemic factors affect bank fundamentals Their relevance increases with time The relevance of fundamentals decreases with time Not just the effect of stale news 25
Conclusions Market discipline as traditionally defined may be overshadowed by the incidence of systemic risk Traditional bank fundamentals may not fully capture systemic risk exposure, materialized prior to crises Depositors may react to systemic risks due to factors unrelated to bank idiosyncratic risk or related to nontraditional fundamentals Systemic risk may affect (future) traditional bank fundamentals, plus available bank-specific information is less informative 26
Conclusions Traditional market discipline is less valuable when market reaction is driven by systemic factors exogenous to bank managers Exposure to systemic risk needs to be added to standard fundamentals (and regulated) Transparent crisis resolution procedures may help reduce dual agency problems inherent to crises Need to limit incidence of systemic risk in emerging economies 27
End 28