liquidity risk and interbank markets - q group · bootstrapping techniques 26. ... liquidity risk...
TRANSCRIPT
1
Liquidity Riskand Interbank Markets
Disclaimer: The views in this presentation are those of the Authors and do not necessarily represent those of the IMF or IMF policy.
Brenda Gonzalez-HermosilloMIT Sloan School of Management
Agenda: Liquidity Risk and Interbank Markets
I. Recent Trends
II. Transmission of Liquidity Spilloversduring the early stages of the crisis
Data and Methodology
III. Interbank Market Rates and Variance Decomposition
IV. Concluding Remarks
2
Agenda: Liquidity Risk and Interbank Markets
I. Recent Trends
II. Transmission of Liquidity Spilloversduring the early stages of the crisis
Data and Methodology
III. Interbank Market Rates and Variance Decomposition
IV. Concluding Remarks
3
Recent Renewed Pressures on Funding Markets
Sources: Bloomberg L.P.; and IMF staff estimates.
1/Forward USD LIBOR minus OIS spread based on December 2010 contract.
2/Spread between 3-month euro/U.S. dollar forex swap and 3-month U.S. overnight index swaps.
3/Spread between 3-month USD LIBOR and the 3-month U.S. overnight interest index swap in basis points.
5
I. Recent Pressures Triggered by the Cristallisation of Sovereign Risks in Europe and Spillover Effects
Central bank liquidity support
0
1
2
3
4
5
6
7
Jun-07 Oct-07 Feb-08 Jun-08 Oct-08 Feb-09 Jun-09 Oct-09 Feb-10
Central Bank Key Policy and Overnight Money Market Rates(In percent)
United Kingdom
Euro area
United States
Source: Bloomberg L.P.
Note: Central bank key policy rates are the following: for the United States, Federal funds target rate; for the United Kingdom, Bank of England's official bank rate; and for euro area, main refinancing operation minimum bid rate.Overnight money market rates are the following: for the United States, Federal funds effective rate; for the United Kingdom, Sterling overnight interbank average (SONIA); and for euro area, Euro overnight interbank average (EURONIA).
7
(10‐yr sovereign swap spreads, percent)
‐1
0
1
2
3
4
Jul‐07 Jan‐08 Jul‐08 Jan‐09 Jul‐09 Jan‐10
Germany France Italy Spain Netherlands
Belgium Austria Greece Ireland Portugal
I.
Financial Crisis Buildup
II.
Systemic
Outbreak
III.
Systemic
Response
IV.
Sovereign
Risk
8
The Four Phases of the Crisis
Source: IMF GFSR, April 2010
9
Sovereign Risk Spillovers Have Emerged as a Threat to Financial Stability
Country level fiscal fundamentals/
vulnerabilities
Financial system fragilities
Sovereign spillovers Funding strains
Liquidity strains during the crisis….
10
In perspective…liquidity strains since the peak of the crisis
Sources: Bloomberg L.P.; and IMF staff estimates.
1/Forward USD LIBOR minus OIS spread based on December 2010 contract.
2/Spread between 3-month euro/U.S. dollar forex swap and 3-month U.S. overnight index swaps.
3/Spread between 3-month USD LIBOR and the 3-month U.S. overnight interest index swap in basis points.
Liquidity strains during the crisis….
0
50
100
150
200
250
300
350
400
Jun-07 Oct-07 Feb-08 Jun-08 Oct-08 Feb-09 Jun-09 Oct-09 Feb-10
Three-Month LIBOR to Overnight Index Swap Spreads(In basis points)
United StatesUnited KingdomEuro area
Source: Bloomberg L.P.
12
Sources: Bloomberg L.P.1/Spread between yields on 90‐day U.S. asset‐backed commercial paper and on the 3‐month U.S. Treasury bill.2/The unweighted daily average of the five‐year credit default swaps for the following institutions: Morgan Stanley, Merrill Lynch, Goldman Sachs, JPMorgan, Deutsche Bank, Bank of America, Citigroup, Barclays, Credit Suisse, and UBS.3/Spread between 3‐month USD LIBOR and 3‐month overnight index swap.
Liquidity strains during the crisis...
Liquidity strains during the crisis...
Agenda: Liquidity Risk and Interbank Markets
I. Recent Trends
II. Transmission of Liquidity Spilloversduring the early stages of the crisis
Data and Methodology
III. Interbank Market Rates and Variance Decomposition
IV. Concluding Remarks
15
II. Transmission of Liquidity Spillovers *
ABCP Funding Liquidity• The financial crisis of 2007 originated from the shock to the U.S.
subprime mortgage market, but this trigger rapidly transmitted to the ABCP market.
• Funding illiquidity experienced by SIVs and conduits as investors became unwilling to roll over the corresponding ABCP.
• This idiosyncratic ABCP funding liquidity shock reflected increasing uncertainty with regard to the value of the underlying securities.
Bank Funding Liquidity• The SIVs began calling on the contingent credit lines from the
sponsoring banks, and the balance sheets of those financial institutions were strained by the reabsorption of the SIVs.
• As a result, the level of interbanking lending declined both for reasons of liquidity and credit risk, leading to higher LIBOR spreads.
16
(*) Nathaniel Frank, Brenda Gonzalez-Hermosillo and Heiko Hesse (2008 a, 2008b, 2010).
Transmission of Liquidity Spillovers
Market Liquidity• Uncertainty with respect to overall market conditions led investors to
increase their demand for the safest and most liquid of all assets.
• Flight to transparency.
Volatility• Financial markets more generally showed signs of stress and
uncertainty.
• Increasing both the volatility of markets, including equity markets.
17
Transmission of Liquidity Spillovers
Credit Default Risk• The crisis also brought to the forefront concerns about the soundness of
a number of financial institutions:– Financial institutions saw a decline in the values of securitized mortgages
and other asset‐backed securities, which resulted in extensive write‐downs.– Higher money market rates increased funding costs.
Liquidity Spirals• Interrelation between market and funding liquidity.• Also account for cyclical effects between market and funding liquidity.
18
Agenda: Liquidity Risk and Interbank Markets
I. Recent Trends
II. Transmission of Liquidity Spilloversduring the early stages of the crisis
Data and Methodology
III. Interbank Market Rates and Variance Decomposition
IV. Concluding Remarks
19
II. Dataand Methodology
• ABCP Spread: 3‐months ABCP yield— 3‐months U.S. Treasury Bill
• Libor Spread: 3‐months U.S. interbank Libor rate — overnight index swap
• Spread between 5‐year on‐the run vs. off‐the‐run U.S. Treasury bonds
• Stock Market Returns: S&P 500, change in returns (volatility)
• Credit default swaps spreads for a sample of representative Large Financial Complex Institutions (LFCIs)
20
The arguments above motivate the variable selection and support the analysis of the transmission of liquidity shocks across five different markets during the early stages of the crisis:
II. Data and Methodology
21
II. Data and Methodology
22
II. Data and Methodology
23
II. Data and Methodology
24
II. Data and Methodology
-15
-10
-5
0
5
10
15
1/3/2006 3/6/2006 5/5/2006 7/6/2006 9/6/2006 11/7/2006 1/8/2007 3/9/2007 5/10/2007 7/11/2007 9/11/2007 11/12/2007-15
-10
-5
0
5
10
15
Returns
CDS
United States: Stock Market Returns and CDS for LCFIs(First difference; in basis points)
Sources: Bloomberg and IMF staff estimates.25
II. Data and Methodology
• Dynamic Conditional Correlation (DCC) model by Engle (2002) which allows for time‐varying correlations. This provides a generalization of the Constant Conditional Correlation (CCC) model by Bollerslev (1990).
• Multivariate DCC GARCH framework. Correlation dynamics.
• Because of a break across all markets, we amend the DCC model to account for a structural break.
• Model in first differences to account for the nonstationarity of the variables in the crisis period.
• Confidence intervals constructed using either Monte Carlo or bootstrapping techniques
26
Methodology
• Sample Period: January 3, 2003‐ January 9, 2008.
• These econometric techniques allow us to analyze the comovement of markets by inferring the correlations of the changes in the spreads discussed above
– which in turn is essential in understanding whether the recent episode of financial distress has become systemic.
27
1. Before the subprime crisis there is only limited correlation between markets.
2. During the crisis period, correlations become more important and bigger in magnitude.
3. Find a statistically significant break in the correlation structure across markets.
28
Main Findings: Transmission of Liquidity Spillovers *
(*) Frank, Nathaniel, Brenda Gonzalez-Hermosillo and Heiko Hesse
• “Transmission of Liquidity Shocks” 2010, in Robert W. Kolb (ed.), Lessons from the Financial Crisis: Causes, Consequences, and Our Economic Future, Hoboken, NJ: John Wiley & Sons, Inc.
• “The Transmission of Liquidity Shocks during the Crisis” 2008b, in Central Banking, Vol. 19, No. 1.
• "Global Transmission of Liquidity Shocks during the 2007 Supbrime Crisis” 2008a, IMF Working Paper WP/08/200.
Main Findings: Transmission of Liquidity Spillovers *
4. Find evidence for more pronounced interaction between market and funding liquidity.
5. Solvency aspects became increasingly important during the crisis.
6. Regarding cross‐border effects, the spillovers occur betweenthe U.S. and international money markets.
7. Stock markets and bond spreads of key emerging marketsare also affected by funding and market liquidity shocks.
29
30
U.S. Model
31
International Model
32
Emerging Market Model
Agenda: Liquidity Risk and Interbank Markets
I. Recent Trends
II. Transmission of Liquidity Spilloversduring the early stages of the crisis
Data and Methodology
III. Interbank Market Rates and Variance Decomposition
IV. Concluding Remarks
33
III. Interbank Market Rates
• The ‘interbank’ in LIBOR and Euribor is somewhat a misnomer.
• Term funds may not be associated with any trades, but a reference interbank rate is still quoted.
• Few, if any, of the banks that contribute to the rate fixings lend term funds to other banks.
• Most unsecured funding comes from non‐bank sources.
• However, interbank rate is an important benchmark for certain loans, including trade finance.
Interbank Market RatesStructured VAR Model (*)
Variance Decomposition of LIBOR/Euribor minus OIS Spread
Structured VAR Model (In percent) *
* Gonzalez-Hermosillo, Brenda, Vance Martin and Miguel Segoviano (forthcoming); Data Jan 1, 2004 to May 28, 2008
Variance Decomposition of LIBOR/Euribor minus OIS SpreadStructured VAR Model (In percent)
Structured VAR Model: Variance Decomposition of 3-month US dollar LIBOR Minus Overnight Index Swap (OIS) Spread
Decomposition of Spread Between Three-Month U.S. Dollar LIBOR and Overnight Index
Swaps (1.1)(In percent)
Sources: Bloomberg L.P.; and IMF staff estimates.
1/Implied volatility from S&P 500 equity index.
2/Lehman Brothers swaption volatility index. Implied volatility of interest rate swaption with maturities ranging from 1 month to 6 months.
Decomposition of Spread Between Three-Month U.S. Dollar LIBOR and Overnight Index
Swaps (1.2)(In percent)
Sources: Bloomberg L.P.; and IMF staff estimates.
3/Five-year on-the-run/off-the-run U.S. treasury note spread.
4/Spread between 3-month euro/U.S. dollar fore swap and 3-month U.S. overnight index swaps.
Decomposition of Spread Between Three-Month U.S. Dollar LIBOR and Overnight Index
Swaps (1.3)(In percent)
Sources: Bloomberg L.P.; and IMF staff estimates.
5/Spread between the yields on 3-month U.S. agency repo and 3-month U.S. treasury repo.
6/Joint probability of distress of selected banks participating in U.S. dollar LIBOR fixing.
Agenda: Liquidity Risk and Interbank Markets
I. Recent Trends
II. Transmission of Liquidity Spilloversduring the early stages of the crisis
Data and Methodology
III. Interbank Market Rates and Variance Decomposition
IV. Concluding Remarks
42
Concluding Remarks
• Financial market innovation has contributed to cross‐border and cross‐market interdependencies
• Interdependencies have significant benefits but...entail risks that have not been fully considered by financial regulators and institutions.
complicate the assessment of liquidity management, risk management and counterparty risk, and policy measures.
43
Concluding Remarks
• Need to assess these interlinkages and, more generally, the mechanisms that may lead to systemic liquidity pressures.– Externalities may have an indirect impact on third parties and create liquidity spirals.
• Methodologies include:– Those that rely primarily on institutional data (e.g. direct network analysis).
– Those that draw inference from market data (e.g. default risk models, co‐risk analysis, etc.)
– Challenge is how to account for indirect and state‐contingent liquidity risks.
44
New Regulatory Landscape – Systemic Liquidity?
• Capital not enough, systemic liquidity risk also needs to be addressed. – Maturity transformation and insufficient stock of highly liquid assets is part of the issue (rollover risks of short‐term liabilities), but not all.
– Need a holistic “measure” of systemic risk.– Need to address non‐bank and non‐deposit sources of systemic liquidity risks.
• Where would a “liquidity charge” be stored? When should be used?– A private or public “insurance” pool?– Central bank(s)?
Key ReferencesInternational Monetary Fund 2008, Global Financial Stability Report, April and October.
Frank, Nathaniel, González‐Hermosillo, Brenda and Heiko Hesse, 2009a, “Transmission of Liquidity Shocks” 2010, in Robert W. Kolb (ed.), Lessons from the Financial Crisis: Causes, Consequences, and Our Economic Future, Hoboken, NJ: John Wiley & Sons, Inc.
Frank, Nathaniel, González‐Hermosillo, Brenda and Heiko Hesse, 2008b, “The Transmission of Liquidity Shocks during the Crisis” 2008b, in Central Banking, Vol. 19, No. 1.
Frank, Nathaniel, González‐Hermosillo, Brenda and Heiko Hesse, 2008a ,"Global Transmission of Liquidity Shocks during the 2007 Supbrime Crisis”, IMF Working Paper WP/08/200.
González‐Hermosillo, Brenda, Miguel Segoviano and Vance Martin, forthcoming, “Frenzy in Interbank Lending Spreads during the Crisis of 2007‐09: A Matter of Liquidity or Solvency?”, IMF Working Paper, International Monetary Fund.
47
Liquidity Riskand Interbank Markets
Disclaimer: The views in this presentation are those of the Authors and do not necessarily represent those of the IMF or IMF policy.
Brenda Gonzalez-HermosilloMIT Sloan School of Management