Are you sure?
This action might not be possible to undo. Are you sure you want to continue?
PRMIA Credit Risk Forum
Evan Picoult Citi Risk Oversight
Evan Picoult, February, 2008
1
Table of Contents
BASIC QUESTION: WHAT ARE THE CONSEQUENCES OF CREDIT EXPOSURE DEPENDING ON THE UNCERTAIN POTENTIAL FUTURE STATE OF MARKET RATES? 1. 2. 3. 4. 5. 6. 7. 8 9 Measuring counterparty credit exposure: Portfolio Simulation Economic capital for loans Economic capital for counterparty credit risk: Default only perspective Economic capital: Economic loss perspective and the CVA for counterparty risk Mitigating counterparty credit risk Credit default swap (CDS) as a hedge of counterparty credit risk Contingent credit default swaps (CCDS) as a hedge of counterparty credit risk Decomposing and dynamically hedging counterparty credit risk Summary of methods to hedge counterparty credit risk 3 10 14 23 32 35 38 47 51
Evan Picoult, February, 2008
2
1. Measuring counterparty credit exposure: Portfolio Simulation
Evan Picoult, February, 2008
3
Counterparty credit exposure – single transaction
The potential exposure profile over time of a single OTC derivative is uncertain. It is contingent on the path market rates follow over time.
Example 1: Forward FX, We buy GBP and sell US$ for settlement in two years at 1.5000 US$/GBP. Random path of forward FX rate for a fixed settlement date, over life of forward transaction in scenario 1. Profile of market value of forward FX transaction over its life, for scenario 1. Exposure Profile of transaction for scenario 1. We only have exposure when the contract has a positive value to us.
Random Scenario 1 for Forward FX Rate
1.750
Forward FX Replacement Cost for Scenario 1 Replacement Cost (% Notional)
20% 15% 10% 5% 0% 5% 10% 15% 20%
Forward FX Exposure Under Scenario 1 Potential Exposure (% Notional)
20%
Forward Exchange Rate
1.625
15%
1.500
10%
5%
1.375
0% 0 3 6 9 12 15 18 21 24
1.250
0
3
6
9 12 Time (months)
15
18
21
24
0
3
6
9
12
15
18
21
24
Time (Months)
Time (months)
Evan Picoult, February, 2008
4
Counterparty credit exposure
Example 2: Three year Fixed/Floating LIBOR Interest Rate Swap
Random Rate Scenario
Potential Exposure (% Notional)
Profile of IR Swap Value
Potential Exposure (% Notional)
IR Swap Exposure Profile
8% Interest Rate 7% 6% 5% 4% 3% 0 6 12 18 24 30 36 Time (months)
6% 4% 2% 0% 2% 0 6 12 18 24 30 36
6% 4% 2% 0% 0 6 12 18 24 30 36
Time (Months)
Time (Months)
If we simulate thousands of paths of the market we can represent the potential exposure profile of a contract at a specified confidence level:
Forward FX Exposure Profiles at Three Confidence Levels
Exposure Profile (% Notional) 60%
Int. Rate Swap Exposure Profiles at Three Confidence Levels
7% Exposure Profile (% Notional)
99.0% CL Profile. 97.7% CL Profile.
50% 40% 30% 20% 10% 0% 0 3 6 9 12 15 18 21 24 Time (Months)
6% 5% 4% 3% 2% 1% 0% 0 6 12 18 24 30 36 Time (Months)
99% CL Profile 97.7% CL Profile Expected Profile
Expected Profile
Evan Picoult, February, 2008
5
Potential Exposure For A Counterparty With Multiple Transactions
Two methods for measuring counterparty exposure (CE) of counterparty with multiple transactions : Simple “AddOn” method for each transaction CE TRANSACTION = = CE CP PORTFOLIO = Current MTM Current MTM + + “Worst case” potential increase in value Notional Principal * Credit exposure factor Potential increase in value per unit of notional principal given transaction’s features.
Σ CE TRANSACTION
Portfolio simulation method: CE
CP PORTFOLIO
=
THE EXPOSURE PROFILE OF COUNTERPARTY
COUNTERPARTY EXPOSURE PROFILE
150
POTENTIAL REPLACEMENT COST ($mm)
125 100 75 50 25 0 0 6 12 18 24 30 36 42 48 54 60
Potential exposure to a counterparty, at a high C.L., over lifetime of transactions with counterparty. Assumes:  No additional transactions  Contractual cash flows set and settle over time.  All legally enforceable risk mitigant agreements are taken into account.
TIME (months)
Evan Picoult, February, 2008
6
Counterparty Exposure Portfolio Simulation
FX FX OPT DEBT SEC. I.R. DER. EQ. DER.
COMM.
COLLATERAL SYSTEM
DER.
PRODUCT PROCESSOR SYSTEMS
Detailed T&C of Transaction
O
DETAILED CONTRACT TERMS AND CONDITIONS.
Credit Admin
O
TABLES OF LEGAL AGREEMENTS

NETTING MARGIN
COUNTERPARTY CREDIT DATA BASE
O
COLLATERAL
Credit Admin
O
TABLES OF DEFAULT TRANSACTION PROFILES
COUNTERPARTY’S:  TRANSACTION DETAILS.  RISK MITIGANT DATA.
COUNTERPARTY’S: EXPOSURE PROFILE.
CE SERVER (analytical engine)
ANALYTICAL ENGINE
Tables of historical or implied volatilities and correlations
Daily feeds of current market data
MARKET DATA
Evan Picoult, February, 2008
7
General method to simulate counterparty’s exposure profile:
1)
Loop over thousands of paths P.
Simulate a path, p, of market rates over time, M(t)P
 Start with current market rates.  Simulate a scenario (or path) of market rates at many future dates, over many years, using tables of volatilities and correlations.
2)
For simulated path, p, measure the potential market value over time of each transaction with counterparty K.
 Start with feed of transaction details and legal information.  For each simulated scenario, calculate the potential market value of each contract at many future dates, using the contract’s terms and conditions, revaluation formula and the simulated state of the market.
3)
Then for simulated path, p, derive counterparty K’s potential exposure over time
 For each simulated scenario, at each future point in time, transform the
potential market value of each contract into the potential exposure of the portfolio through aggregation rules that take risk mitigants and legal context into account.  i.e. For the counterparty K, for path M(t)P derive Exposure(t)K,P
4)
After simulating thousands of potential paths of market rates, M(t)p Calculate exposure profile of counterparty: the potential exposure at some high confidence level, at a set of forward dates
Evan Picoult, February, 2008
8
Counterparty credit exposure
The potential exposure profile of a counterparty with many transactions is contingent on all the market factors that effect the value of those transactions as well as any legally enforceable risk mitigation agreements, such as netting or margin. It is complex to model because of the need to take into account the effect of legally enforceable risk mitigation agreements like netting and margin. If we simulate thousands of paths of the market we can represent the potential exposure profile of a portfolio of contracts at a specified confidence level:
Potential exposure profile of a counterparty, at two confidence levels, over the lifetime of the transactions with the counterparty: Assumes:
Potential Exposure ($MM)
125 100 75 50 25 0 0 6 12 18 24 30 36 42 48 54 60
A Counterparty's Exposure Profile
150
• No additional transactions • Contractual cash flows are set and settle over time. • All legally enforceable risk mitigation agreements are taken into account
99%CL Expected Positive Exposure (EPE)
Time (Months) ==>
Evan Picoult, February, 2008
9
2. Economic Capital for loans
Evan Picoult, February, 2008
10
EC Definition
Economic Capital (also called “Economic Risk Capital” or “Risk Capital”) is a measure of risk. Risk in this context means the potential unexpected loss of economic value over one year, calculated at a very high confidence level (99.97% CL). Thus EC measures risk from an insolvency or debt holders perspective (potential loss of value) rather than from an equity investment perspective (undiversified volatility of returns). Here is an example of EC for a loan portfolio:
Probability Distribution of Potential Credit Loss for a Portfolio of Many Obligors
3.0%
Probability of Credit Loss
Loss at high CL
Expected Loss
2.5% 2.0% 1.5% 1.0% 0.5% 0.0% 160
Economic capital
= Unexpected Loss = Loss at very high CL – Expected loss.
Economic Capital
140
120
100
80
60
40
20
0
Potential Credit Loss ($mm)
Expected loss should be covered by reserves and/or pricing.
• Finally, need to allocate total EC to the EC per loan as a function of risk characteristics of obligor, loan facility and risk concentration
Evan Picoult, February, 2008
11
Economic Loss  Loan Portfolio  Default Only Analysis
ASSUME SOURCE OF CREDIT RISK IS DEFAULT AND RECOVERY ONLY.
• Factors needed to simulate total loss distribution:
 Credit exposure per obligor  Probability distribution of exposure at default, for contingent credit.  Probability of default and correlations of probability of default  Probability distribution of loss given default (LGD) (i.e. 1 – recovery%).
• A robust method will model and capture the relative degree of risk diversification or risk concentration in the portfolio. • Also need a method for allocating the total EC across all obligors to each loan facility, as a function of risk characteristics of obligor and loan facility. • There are several very different ways of modeling the potential loss distribution due to default and recovery and of allocating total EC to each credit facility.
Evan Picoult, February, 2008
12
Economic Loss  Loan Portfolio – Economic Loss Analysis
ASSUME SOURCE OF CREDIT RISK IS ECONOMIC LOSS • Factors needed to simulate loss distribution:
 Credit exposure per obligor
Components of default only perspective
 Probability distribution of exposure at default, for contingent credit.  Probability of default and correlations of probability of default.  Probability distribution of loss given default (LGD).
Component of long term simulation of obligor’s spread
 Probability distribution of migration of PD (internal risk rating).  Volatilities and correlations of change in spread, given rating.  Potential change in idiosyncratic spread.
• There are several very different ways of modeling the potential loss distribution due to economic loss.
Evan Picoult, February, 2008
13
3. Economic Capital for counterparty credit risk: Default only perspective
Evan Picoult, February, 2008
14
Overview Of Analysis
Full Simulation of Economic Capital: Joint simulation of exposure, default and recovery
nd t a ry os ul e fa ov ari De rec cen s
Default/Recovery Scenario J ... Default/Recovery Scenario 2 Default/Recovery Scenario 1
SIMULATION OF COUNTERPARTY EXPOSURE PROFILE
Set of Transactions Set of Legally Enforceable Risk Mitigation Agreements Simulated paths of market Path 1 Path 2 ... Path M Exposure Profile of Portfolio of Transactions with Counterparty at x%CL
Counterparty 1
Counterparty 2
...
Counterparty N
Potential Loss
Distribution
Exposure profile of CP 1 for path 1 ... ...
Exposure profile of CP 2 for path 1 ... ...
... ... ...
Exposure profile of CP N for path 1 ... ...
Exposure Profile for CP 1 at x%CL
Exposure Profile of CP 2 at x%CL
...
Exposure Profile of CP N at x% CL
Evan Picoult, February, 2008
15 Simulation of exposure profile
Economic Loss  Counterparty Risk  Default Only Analysis
– EC For Counterparty Risk vs. EC For Loan Risk From A Default Only Perspective.
LOAN PORTFOLIO Default and recovery Variable exposure Intercounterparty portfolio w.r.t. default. TYPES OF DIVERSIFICATION BENEFITS Intercounterparty portfolio w.r.t. exposure Intracounterparty portfolio w.r.t. exposure √ √ COUNTERPARTY CREDIT RISK √ √ √ √
Decreases Risk Increases Risk
DRIVERS OF WIDTH OF LOSS DISTRIBUTION
√
– Full simulation of EC from a default only perspective:
For a given path of the market over time, the risk free value of each contract and the corresponding conditional exposure of each counterparty can be fully specified. For each path we can therefore simulate thousands of scenarios of default and recoveries, exactly as we would for a loan portfolio from a default only perspective. We can then loop over thousands of potential scenarios of changes in market rates.
Evan Picoult, February, 2008
16
Economic Loss  Counterparty Risk  Default Only Analysis
EC BY FULL SIMULATION: GENERAL METHOD, FIVE STEPS
1) SIMULATE A PATH, P, OF MARKET RATES OVER TIME M(t)P
Same as for Exposure Profile. Loop over thousands of paths P.
2)
FOR SIMULATED PATH P, MEASURE THE POTENTIAL MARKET VALUE OVER TIME OF EACH Same as for Exposure Profile. TRANSACTION WITH EACH COUNTERPARTY K.
3)
FOR SIMLUATED PATH P, DERIVE COUNTERPARTY K’S POTENTIAL EXPOSURE OVER TIME. i.e. For each counterparty K, for path M(t)P derive Exposure(t)K,P Same as for Exposure Profile.
Loop over thousands of scenarios of default and recovery.
4)
USING THE SET OF EXPOSURE PROFILES {Exposure(t)K,P } FOR EACH AND EVERY COUNTERPARTY K, GENERATED BY MARKET PATH P: CALCULATE THE POTENTIAL LOSS DISTRIBUTION BY SIMULATING THOUSANDS OF SCENARIOS OF DEFAULT AND RECOVERY FOR THE SET OF COUNTERPARTIES K.
5)
AFTER SIMULATING THOUSANDS OF POTENTIAL PATHS OF MARKET RATES, M(t)P CALCULATE FULL LOSS DISTRIBUTION AND DERIVE THE FULL SIMULATION ECONOMIC CAPITAL FOR COUNTERPARTY RISK.
Evan Picoult, February, 2008
17
Economic Loss  Counterparty Risk  Default Only Analysis
Is it possible to define a “loan equivalent” for counterparty exposure?
– A “Loan Equivalent” is the fixed exposure profile, per counterparty, that would generate the same economic capital as obtained by full simulation of changes in market rates, default and recovery.
How good an approximation is the expected exposure profile as a loan equivalent? In what context, if any, is it a good approximation?
Note:
– There is no actual scenario of market rates that corresponds to each counterparty having an exposure equal to its average potential positive exposure. In fact, for a large derivative trader there will be no scenario for which every counterparty will have positive exposure at the same time. Nonetheless, the loan equivalent is a valid and very useful method to assess the total Economic Capital due to counterparty risk, across all counterparties.
Why does question arise?
– In Basel II: Risk Weighted Asset = EAD * RW(PD, LGD, M) where RW is the risk weight function. Therefore to calculate the RW for counterparty risk we need to measure the EAD for counterparty risk (i.e. the loan equivalent) and M (the effective maturity). – Internally, transactors need an easy method for evaluating EC and return on EC. Therefore they need to have the loan equivalent (EAD in Basel II) and the effective maturity (M in Basel II) as parameters to input into EC tables or EC analytic functions (as in Basel II).
Evan Picoult, February, 2008
18
Economic Loss  Counterparty Risk  Default Only Analysis
EC using expected positive exposure profile as a loan equivalent: 1) Calculate the expected positive exposure profile, EPE(t)k , of each counterparty K
Loop over thousands of paths P.
a) Simulate a path, P, of market rates over time M(t)P b) For simulated path, P, measure the simulated market value over time of each transaction with counterparty K. c) For simulated path, P, derive counterparty exposure over time. i.e. For the counterparty K, for path M(t)P , derive Exposure(t)K,P
d) Derive each counterparty’s expected positive exposure profile EPE(t)K
Loop over thousands of default scenarios.
2)
Use the set of expected positive exposure profiles {EPE (t)k} for every counterparty
Calculate the potential loss distribution by simulating thousands of scenarios of default and recovery for the set of counterparties K.
3)
Calculate the economic capital from the loss distribution derived from each counterpartry’s expected exposure profile.
19
Evan Picoult, February, 2008
Economic Loss  Counterparty Risk  Default Only Analysis
EC using EPE * α as a loan equivalent: Shortcut to full simulation of exposure, default and recovery: EAD = α * EPE
– Define a portfolio parameter alpha: α
α
=
Economic Capital Economic Capital
FULL_SIM; DEFAULT ONLY
FIXED_EPE_ SIM; DEFAULT ONLY
– Where the denominator of the ratio is defined as the economic capital measured by assuming that the exposure profile of each counterparty can be expressed as a constant exposure whose magnitude is EPEK , the average of the EPEK(T) profile over a one year horizon. – EPEK = Σ EPEK(tj) / (tj  tj1) – For a large trading desk, α ≅ 1.1 to 1.2 (e.g. see Canabarro, Picoult, Wilde; Risk, Sept. 2003)
For each counterparty, to a very good approximation, the economic capital per counterparty can be calculated by treating the counterparty exposure as if it were a fixed exposure of magnitude:
– EAD (Exposure at Default) = α * EPE where EPE is the average of the EPE(t) profile over one year. – Note: In a default only analysis, the effective maturity of the exposure is one year.
Evan Picoult, February, 2008
20
Economic Loss  Counterparty Risk  Default Only Analysis
ISDA TESTS
Create test portfolios and calculate α as a function of the characteristics of the portfolio:
Effective number of counterparties Effective number of market factors Probability of default of counterparties Correlation of default. Initial MTM of counterparties’ portfolios. Other factors.
Initial Proposal: …………………………………………Evan Picoult, Citigroup Simulations of Stylized Portfolios:…………………..Eduardo Canabarro, GS (now Morgan Stanley) Analytical Calculations:………………………………..Tom Wilde, CSFB Measurement of a for Real Portfolios:………………Several Firms α ≈ 1.1
SUMMARY CONCLUSION:
See: 
For Large Market Makers
ISDA web site, Papers on Counterparty Risk to Basel Committee, June 2003 Risk Magazine, (Canabarro, Picoult, Wilde) September, 2003
Later, additional calculations were done: For “general wrong way risk”: For Large Market Maker α ≈ 1.2
but some banks have portfolios with mostly “general right way risk”.
Evan Picoult, February, 2008
21
Economic Loss  Counterparty Risk  Default Only Analysis
α across various portfolio characteristics
Data slide originally from Eduardo Canabarro, now at Morgan Stanley.
Alpha
Current exposure Number of risk factors Number of ctpt'ies Prob of default MCarlo (Eduardo) Analytic (Tom)
Alpha with WrongWay Risk (Rsq = 20% )
M Carlo (Eduardo) MCarlo (Marcus) Analytic (Tom)
Correlation
Percentile
Base case 22% 1.36 3 Sensitivity to credit correlation 0% 1.36 3 1.36 3 12% 24% 1.36 3 50% 1.36 3 Sensitivity to current exposure level 22% 0 3 22% 1 3 22% 2 3 22% 3 3 Sensitivity to number of market risk factors 22% 1.36 1 22% 1.36 5 22% 1.36 10 22% 1.36 50 Sensitivity to number of counterparties
22% 22% 22% 22% 22% 22% 22% 22% 22% 22% 1.36 1.36 1.36 1.36 1.36 1.36 1.36 1.36 1.36 1.36 3 3 3 3 3 3 3 3 3 3
200
200 200 200 200 200 200 200 200 200 200 200 200
0.30%
0.30% 0.30% 0.30% 0.30% 0.30% 0.30% 0.30% 0.30% 0.30% 0.30% 0.30% 0.30% 0.30% 0.30% 0.30% 0.30%
99.90%
99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90% 99.90%
1.09 1.43 1.21 1.08 1.02 1.35 1.14 1.05 1.03 1.10 1.08 1.07 1.07 1.26 1.22 1.10 1.04 1.17 1.07 1.06 1.05 1.07 1.10
1.08
1.12 1.43 1.23 1.10 1.05 1.69 1.24 1.02 1.01 1.42 1.16 1.09 1.06 1.35 1.35 1.23 1.12 1.22 1.09 1.08 1.07 1.04 1.08
1.21 1.43 1.29 1.17 1.09 1.47 1.28 1.12 1.07 1.42 1.15 1.32 1.28 1.28 1.11 1.31 1.18 1.13 1.13 1.10 1.15
1.15
1.15 1.07 1.02 1.33 1.12 1.04 1.02 1.09 1.08 1.31 1.20 1.13 1.04 1.12 1.06 1.05 1.04 1.10 1.09
1.22 1.14 1.09 1.58 1.25 1.07 1.03 1.38 1.12 1.09 1.08 1.39 1.28 1.20 1.11 1.20 1.14 1.12 1.12 1.13 1.13
Key to names • Eduardo = Eduardo Canabarro was at Goldman now at Lehman • Tom = Tom Wilde CSFB • Marcus = Marcus Fleck Dresdner Bank
20 50 100 500
200 200 200 200 200 200
Sensitivity to probability of default 0.10% 0.50% 1.00% 5.00%
0.30% 0.30%
Sensitivity to confidence level 99.00% 99.50%
Evan Picoult, February, 2008
22
4. Economic Capital for counterparty credit risk: Economic Loss Perspective and the CVA
Evan Picoult, February, 2008
23
Economic Loss  Counterparty Risk  Economic Loss Analysis
First Question: – What should be the effect of credit spreads / risk rating on derivative valuation? – If all derivatives are (and should be) markedtomarket by discounting expected future cash flows at LIBOR bid/offer midpoint, then valuation would be:
• Independent of counterparty’s risk rating and • Independent of counterparty’s credit spread.
– If that were the case, changes in risk rating or spreads would not cause a change in economic value. Default only and economic loss analysis would be the same.
Evan Picoult, February, 2008
24
Economic Loss  Counterparty Risk  Economic Loss Analysis
What is the credit risk premium for OTC derivatives? Let us first review the credit risk premium for a loan:
Bond/Loan Value ∴ Risk Premium Loan/bond = PV of cash flows discounted at Risk Free Rate – RISK Premiumloan/Bond = PV of cash flows discounted at (Risk Free Rate + SPREAD) ≅ PVRisk Free * Duration * Average SpreadLoan/bond (from simple math)
Let us try to apply the same logic to calculate the credit risk premium of a derivative:
– Derivative Value = PV of cash flows discounted at Risk Free Rate – Risk Premium
However, the context for ascertaining the credit risk premium of derivatives is material different than it is for loans because of the different nature of their respective credit exposures. We need to:
– Perform a portfolio analysis of exposure – Take into account the uncertain future exposure – Evaluate significance, if any, of bilateral nature of exposure In general, at any future date • • Obligor could owe us (asset) or We could owe obligor (liability) Method for calculating risk premium may depend on the purpose of the calculation:  Pricing  Cost of credit
Therefore how should we take the potential bilateral nature of exposure into account?
Evan Picoult, February, 2008
25
Economic Loss  Counterparty Risk  Economic Loss Analysis
Defining a unilateral and a bilateral CVA
Let us call the credit risk premium of the counterparty’s portfolio its CVA.
CVA = Credit value adjustment for counterparty’s credit risk
Therefore, the market value of derivative portfolio with a counterparty is:
= Σ MARKET VALUE (discounted at risk free rate)  COUNTERPARTY RISK PREMIUM = Σ MARKET VALUE (discounted at risk free rate)  CVA
Calculated on a portfolio basis, taking into account potential future exposure
Measuring the cva of a counterparty: Modification of a proposal by Bollier & Sorensen. Two perspectives on cva: a unilateral and a bilateral perspective.
Unilateral CVA: CVACOUNTERPARTY K, UNILATERAL Bilateral CVA: CVACOUNTERPARTY K, BILATERAL = CVA+CNTPY K
Credit premium of own firm’s expected asset from derivatives
= CVA+CNTPY K  CVA– CNTPY K
Credit premium of CP’s expected asset from derivatives.
Evan Picoult, February, 2008
26
Economic Loss  Counterparty Risk  Economic Loss Analysis
Defining a unilateral and a bilateral CVA
Market Value CP Portfolio = Σ MVCP Portfolio (risk free)  CVA CP Portfolio CVA CP Portfolio_Unilateral
Expected amount CP will owe to own firm.
A Counterparty's Expected Positive Exposure Profile
75 Potential Exposure ($MM) 60 45 30 15 0 0 6 12 18 24 30 36 42 48 54 60 Time (Months)
= CVA+ CNTPY = CVA+ CNTPY  CVA
CNTPY
CVA CP Portfolio_Bilateral CVA+CNTPY K
J
(Credit premium of own’s firm potential asset due to CP K)
= ∑ ( Expected Exposure + ,J * Forward CP Spread K ,J * ∆ t J * df J ) K
Calculate and sum over each forward period J, for CP K.
Expected amount own firm will owe to CP.
A Counterparty's Expected Negative Exposure Profile
75 Potential Exposure ($MM) 60 45 30 15 0 0 6 12 18 24 30 36 42 48 54 60 Time (Months)
CVA
CNTPY K
(Credit premium of CP K’s potential asset due to own firm)
= ∑ ( Expected Exposure − ,J * Forward Own Firm' s Spread J * ∆ t J * df J ) K
J
Calculate and sum over each forward period J, for CP K
Evan Picoult, February, 2008
27
Economic Loss  Counterparty Risk  Economic Loss Analysis
Implications of a unilateral and a bilateral CVA
Market Value CP Portfolio CVA CP Portfolio_Bilateral CVA CP Portfolio_Unilateral Examples = = = Σ MVCP Portfolio (risk free)  CVA CP Portfolio CVA+ CNTPY  CVA CVA+ CNTPY
CNTPY
(bilateral perspective) (unilateral perspective)
Example 1: • Assume: – Only One Swap With Counterparty
– Counterparty And Own Firm Have Same Risk Rating. – Potential Change in Value Has Symmetric Shape For Pay or Receive Fixed Swaps. (e.g. flat yield curve).
• Consequence for unilateral and bilateral CVA Example 2: • Assume: – Only One Swap With Counterparty
– Counterparty Is BBB And Own Firm Is AA.
• Consequence for unilateral and bilateral CVA
Evan Picoult, February, 2008
28
Economic Loss  Counterparty Risk  Economic Loss Analysis
Deriving Duration, M, for Unilateral CVA: For simplicity, let us define Spread = average spread for counterparty EPE1 yr = Average EPE over one year horizon.
Let us first assume that EPE stays constant but that spreads can change. We have:
Full life of portfolio
CVA
=
Spread * ∆Spread *
∑ EPE
k =1 k =1
k
∆tk df k ∆tk df k
Full life of portfolio
∆CVA =
∑ EPE
k
In analogy to the relationship between the change in the credit spread of a bond and its duration, we can define the effective M for the unilateral CVA by the equation:
∆CVA ≡ ∆Spread * EPE1 year * M
(Similar to: ∆Risk PremiumBond = ∆ SpreadBond * PVRisk Free * Duration)
We therefore have the definition of M, from a unilateral perspective:
Maturity
∑ EPE
M =
k=1 1 Year k=1
k
∆tk df k ∆tk df k
M is simply the ratio of: The area under the fulllifetime discounted EPE curve divided by the area under the 1year discounted EPE curve.
∑ EPE
k
For fuller discussion including M for bilateral CVA, see Evan Picoult and David Lamb (2004) Economic Capital for Counterparty Credit Risk, Chapter in Economic Capital: A Practitioner Guide, London, Risk Books
Evan Picoult, February, 2008
29
Economic Loss  Counterparty Risk  Economic Loss Analysis
In Basel II EAD = α * Effective EPE
Expected Exposure Profile
2.5 Expected Exposure 2 1.5 1 0.5 0 0 3 6 9 12 Time (months) 15 18
What is Effective EPE?
Expected Exposure Profile and the EPE averaged over one year.
Example of expected positive exposure increasing in each forward period.
Expected Exposure Profile
EPE averaged over one year
Expected Exposure Profile
2.5 Expected Exposure 2 1.5 1 0.5 0 0.5 0 3 6 9 12 Time (months) 15 18
Expected Exposure Profile
Expected Exposure Profile and the EPE averaged over one year.
Example of expected positive exposure increasing then decreasing over time.
EPE averaged over one year
As might occur for portfolio with shortdated transactions
Effective Expected Exposure Profile
2.5 Expected Exposure 2 1.5 1 0.5 0 0.5 0 3 6 9 12 Time (months) 15 18
Effective Expected Exposure Profile Expected Exposure Profile Effective EPE averaged over one year
Effective Expected Exposure Profile and the Effective EPE averaged over one year.
Effective Expected positive exposure is defined to never decrease over the first year. Effective EEk = max(Effective EEk1, EEk)
Evan Picoult, February, 2008
30
Economic Loss  Counterparty Risk  Economic Loss Analysis
A Counterparty's Exposure Profile
150 Potential Exposure ($MM) 125 100 75 50 25 0 0 6 12 18 24 30 36 42 48 54 60 Time (Months) ==>
Lifetime area under discounted curve
Remaining life of portfolio
∑ EPE
k=1
k
∆tk df k
= 140.7
A Counterparty's Exposure Profile
150 Potential Exposure ($MM) 125 100 75 50 25 0 0 6 12 18 24 30 36 42 48 54 60 Time (Months) ==>
One year area under discounted curve
1 Year
∑ EPE
k =1
k
∆tk df k
= 75.9 = EPE1 Year
Therefore
M
= = =
(140.7/75.9) yrs 1.85 yrs 22.2 months
≤ 1 yr
For Basel II calculation of M: Need to use Effective EPE over the first year in numerator and denominator rather than EPE
∑
M=
k =1
Effective EPE k
∆t k
df k
Remaining life of portfolio +
∑
k > 1 yr
EPE k df k
∆t k
df k
1 Year k =1
∑ Effective EPE k ∆t k
Evan Picoult, February, 2008
31
5. Mitigating Counterparty Credit Risk
Evan Picoult, February, 2008
32
Growth of OTC Derivative Market
OTC Derivative Market continues to grow exponentially, measured by notional outstanding:
GROWTH OF OTC DERIVATIVE MARKET All Currencies (ISDA) (all dates endofyr, but 2007 is as of midyear)
Equity Derivatives
450,000 NOTIONAL PRINCIPAL ($bn) 400,000 350,000 300,000 250,000 200,000 150,000 100,000 50,000 0
Credit Default Swaps Total IR and Currency 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007
Year End
Because counterparty credit exposure is contingent on market rates:
– Counterparty credit exposure is more difficult to measure than lending exposure. – Counterparty credit risk is more difficult to measure and hedge than lending risk.
Evan Picoult, February, 2008
33
Methods for reducing counterparty credit risk
Comparison of different methods for reducing or hedging counterparty credit risk.
Method Netting
Reduce counterparty exposure
Requirements to be an effective risk mitigant Reduces exposure if counterparty has transactions with offsetting market values: – Oppositely positioned transactions. – Or, potential benefit for a large set of transactions on different, not perfectly correlated, underlying market factors.
None – other than a low minimum threshold.
Costs Low costs to enter into. Moderate cost to build process to accurately calculate portfolio exposure profile. Relatively high infrastructure costs and liquidity impact.
Interbank Usage and Impact Widely used in interbank market. Strong impact
Corporate enduser Usage and Impact Widely used. Usually weak impact.
Margin
Widely used in interbank market. Very strong impact when used.
Rarely used because corporations don’t have infrastructure and don’t want liquidity risk. Big need
Reduce PD
Hedge Counterparty Credit Risk
Not needed when there are margin agreements.
Evan Picoult, February, 2008
34
CVA
Market Value CP Portfolio = Σ MVCP Portfolio (risk free)  CVACP Portfolio
Two choices:
– Credit Risk Perspective CVA is treated as part of credit risk. This bifurcation of counterparty risk into a) market risk and b) credit risk is identical to what occurs under accrual accounting, where the risk of a loan portfolio is bifurcated into a) accrual interest rate risk and b) credit risk. From this perspective hedging counterparty risk is done by buying a CCDS from a third party, in the same way that the credit risk of a loan is hedged by buying a CDS or a Guarantee from a third party – Full Market Risk Perspective: CVA is treated as part of market risk, in the same way that the spread of a bond is treated as part of market risk. Dynamically delta hedging CVA is equivalent to transforming counterparty credit risk to market risk and hedging it like market risk. .
Evan Picoult, February, 2008
35
Hedging counterparty credit risk
Alternative means of hedging counterparty risk (i.e. reducing the PD of counterparty default)
Credit Risk Perspective: Reduce the effective exposure to the original counterparty
– CDS: Credit Default Swap Fixed Notional hedge – CCDS: Contingent Credit Default Swap Variable notional hedge, where notional amount is determined by a referenced derivative.
Market Risk Perspective: Treat CVA as market risk and hedge it.
– Dynamic Delta Hedging: Transform counterparty credit risk into market risk by decomposing and dynamically hedging the CVA (Credit Value Adjustment) of counterparty credit risk. – CCDS: Contingent Credit Default Swap CCDS would have to be analyzed in terms of same factor sensitivities as the CVA
Evan Picoult, February, 2008
36
6. Credit Default Swap (CDS) As Hedge of Counterparty Risk
Evan Picoult, February, 2008
37
CDS
A Credit Default Swap (CDS) is a bilateral contract between two parties to isolate and separately trade the credit risk of at least one third party referenced entity:
– The buyer of credit protection pays a periodic fee to the seller in exchange for a contingent payment, on the occurrence of a specified credit event to the referenced entity. – Upon the trigger of the credit event to the referenced entity, the seller must pay the buyer either:
The par amount of a referenced asset (e.g. defaulted bond) in exchange for its physical delivery to the seller. Or the difference between the par amount and the market value of the referenced asset.
The terms and conditions of the CDS specify:
– – – – – The referenced entity. The referenced asset of the referenced entity and its fixed par amount. The definition of the credit event The tenor of the contract The periodic fee.
Periodic payment
Buyer
Contingent payment
Seller
Payment is triggered by occurrence of credit event to referenced entity, and will be of one of two forms: Either: a) Par amount in exchange for physical delivery of referenced asset Or: b) Difference between par and market value of referenced asset.
Evan Picoult, February, 2008
38
Ineffectiveness of CDS for hedging counterparty credit risk
Example:
– 3yr payfixed/receive floating USD LIBOR IRS; – Notional principal of $100mm
How much notional CDS should one buy? Statistical picture of exposure over time at different confidence levels
Int. Rate Swap Exposure Profiles at Three Confidence Levels
7% Exposure Profile (% Notional) 6% 5% 4% 3% 2% 1% 0% 0 6 12 18 24 30 36 Time (Months)
99% CL Profile 97.7% CL Profile Expected Profile
If the assumptions about future volatility are accurate than at the 99%CL, the CDS notional would be set at $6mm (6%*100mm). But this means one would be over hedged more than 99% of the time. Even for the 99%CL profile, the exposure is at $6mm only for a fraction of the three year period. This is an inefficient hedge.
Evan Picoult, February, 2008
39
7. Contingent Credit Default Swap (CCDS) As Hedge of Counterparty Risk
Evan Picoult, February, 2008
40
CCDS
A Contingent Credit Default Swap (CCDS) is a form of a CDS. Its distinguishing characteristic is that the par amount of the referenced asset that must be delivered to the seller is specified by the market value of a referenced derivative.
– As a consequence, the par amount of the referenced asset is contingent on the state of the market.
Definition:
– A CCDS is a bilateral contract between two parties to isolate and separately trade the Counterparty Credit Risk of at least one third party referenced entity: – The buyer of credit protection pays a periodic fee to the seller in exchange for a contingent payment, on the occurrence of a specified credit event to the referenced entity. – Upon the trigger of the credit event to the referenced entity: The par amount of the referenced asset is set to the current market value of the referenced derivative, discounted at LIBOR flat. The seller must pay the buyer either: The par amount of a referenced asset (e.g. defaulted bond) in exchange for its physical delivery to the seller. Or the difference between the par amount and the market value of the referenced asset.
Note: The specific terms and conditions of CCDS will evolve and broaden as market grows.
Evan Picoult, February, 2008
41
CCDS
The terms and conditions of the CDS specify:
– The referenced entity. – The referenced asset of the referenced entity – The referenced derivative (whose market value determines the par amount of the referenced asset) and the method for valuing the referenced derivative. – The definition of the credit event – The tenor of the contract – The periodic fee.
Evan Picoult, February, 2008
42
CCDS
Example 1: CCDS which fully hedges a derivative.
– Underlying transaction that needs to be hedged:
A $100mm, 5 year IRS with counterparty A. Counterparty A pays LIBOR and receives fixed semiannually
– CCDS hedge with counterparty B
Referenced entity: Referenced asset: Referenced derivative: Credit Event:
Counterparty A Five year fixed rate bond of counterparty A that is pari passu, in the event of default, with an unsecured derivative with counterparty A. $100mm, 5 year fixed/floating LIBOR IRS. Default by A.
Exposure Profile of underlying swap
Ex. 1 Exposure Profile of 5 Yr. Swap
14
=
Profile of Par Amount of CCDS Hedge
Ex. 1 Profile of Par Amount of CCDS Hedge
14
Potential Replacement Cost ($mm)
12 10 8 6 4 2 0 0 12 24 36 48 60
Potential Replacement Cost ($mm)
12 10 8 6 4 2 0 0 12 24 36 48 60
99% CL Profile Profile for a random path. EPE Profile
Time (Months)
Time (Months)
Evan Picoult, February, 2008
43
CCDS
Example 1 continued
Original exposure to Counterparty A
Hedged Exposure
Residual Exposure to Counterparty A
Use either: • Substitution approach
– Substitute PD of seller of CCDS for PD of underlying counterparty. – Allowed under Basel.
• No residual exposure in this example.
a
• If there were residual exposure, it would simply be treated as unhedged exposure to the underlying counterparty.
• Double Default approach
• Calculate joint probability of default of underlying obligor (A) and seller of CCDS (B). • Basel II has its own double default formula that applies in narrow contexts.
Evan Picoult, February, 2008
44
CCDS
Example 2: CCDS which partially hedges a derivative.
– Underlying transaction that needs to be hedged:
A $100mm, 5 year IRS with counterparty A, with 50% extra notional principal the first 2 years. Counterparty A pays LIBOR and receives fixed semiannually
– CCDS hedge with counterparty B
Referenced entity: Referenced asset: Referenced derivative: Credit Event:
Counterparty A Five year fixed rate bond of counterparty A that is pari passu, in the event of default, with an unsecured derivative with A. $100mm, 5 year fixed/floating LIBOR IRS. Default by A.
Example 2: Comparison of 99%CL Exposure Profile of Underlying Swap vs. Par Amount of CCDS Hedge
Ex. 2 Comparison of 99%CL Exposure Profile of Swap vs. CCDS Hedge
14
Potential Replacement Cost ($mm)
Underlying swap
12 10 8 6 4 2 0 0 12 24 36 48 60
CCDS Hedge
In this example, the exposure profile of the referenced derivative (a plain vanilla swap) is not identical to that of the underlying swap during the first 2 years of the 5 year tenor..
Time (Months)
Evan Picoult, February, 2008
45
CCDS
Example 2: CCDS which partially hedges a derivative.
Profile of Par Amount of CCDS Hedge
Ex. 2 Profile of Par Amount of CCDS Hedge
14
Potential Replacement Cost ($mm)
12 10 8 6 4 2 0
Exposure Profile of underlying swap with A
Ex. 2 Exposure Profile of 5 Yr. Swap
Potential Replacement Cost ($mm)
14 12 10 8 6 4 2
0
12
24
36
48
60
99%CL
Time (Months)
Profile of Unhedged Exposure to A
EPE
14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 0 12 24 36 48 60
Ex. 2 Unhedged Exposure Proflie at 99%CL and EPE
Potential Replacement Cost ($mm)
0 0 12 24 36 48 60
Time (Months)
Time (Months)
Evan Picoult, February, 2008
46
CCDS
Example 2 continued
Original exposure to Counterparty A
Hedged Exposure
Residual Exposure to Counterparty A
Use either: • Substitution approach
– Substitute PD of seller of CCDS for PD of underlying counterparty. – Allowed under Basel.
• This example has residual exposure to counterparty A.. • It will simply be treated as unhedged exposure to the underlying counterparty.
a
• Double Default approach
• Calculate joint probability of default of underlying obligor (A) and seller of CCDS (B). • Basel II has its own double default formula that applies in narrow contexts.
Evan Picoult, February, 2008
47
CCDS
Reference Derivative of CCDS will most likely be a plain vanilla, very liquid derivative.
– Transparency in pricing. – Ease of delta hedging the CCDS by the seller of the CCDS.
Questions
– What should be the characteristics of the referenced asset for it to function as an optimal hedge? Answer: The referenced asset should be pari passu with derivative it is hedging, so that the LGD will be the same.
– How effective is the CCDS as a credit risk mitigant under Basel II? Answer: Recognized by US implementation of Basel II, subject to approval by primary supervisor.
– What is the relationship between the CVA (Credit Value Adjustment) of a derivative (or of a portfolio of derivatives with a counterparty) and the value of the CCDS? – What alternative means of hedging are there? Answers in the remaining sections.
Evan Picoult, February, 2008
48
8. Decomposing and dynamically hedging the CVA
Evan Picoult, February, 2008
49
Dynamically Hedging Counterparty Risk
Market Value CP Portfolio = Σ MVCP Portfolio (risk free)  CVACP Portfolio
Gives rise to market risk EC.
Full life of portfolio
Gives rise to counterparty credit risk EC.
CVAUnilateral
= Spread *
∑ EPE
k =1
k
∆tk df k
• For a specific portfolio of U.S. Dollar LIBOR interest rate swaps, the discounted area under the EPE curve, Σ( Σ(EPEk ∆tk dfk), will be a function of the terms and conditions of all the swaps and the term structure and volatility of the U.S. Dollar LIBOR yield curve.
• EPE of counterparty depends on Terms and conditions of contracts, legally enforceable risk mitigant agreements, yield curve and assumed volatilities and correlations. • Represent interest rates by a single variable X and interest rate implied vol by a single variable σ .
2
Full life of portfolio
CVA
= Spread *
∑ EPE
k =1
k
∆tk df k
= Spread * g(Contract T & Cs, Risk Mitigants ; {r(t)k }, {σk }) = Spread * g(Contract T & Cs, Risk Mitigants; xk ,σ k ) = Spread * g(Contract T & Cs, Risk Mitigants; x, σ )
Full life of portfolio
∆CVA
= ∆Spread *
∑ EPE
k =1
k
∆tk df k
δCVA ∂CVA ∂ CVA * ∆Spread * ∆x + . . . * ∆x + Spread * * ∆σ + + Spread * δσ ∂x ∂Spread∂x
Evan Picoult, February, 2008
50
Dynamically Hedging Counterparty Risk
Full life of portfolio
CVA
= Spread *
∑ EPE
k =1
k
∆tk df k
Full life of portfolio
∆CVA
= ∆Spread *
∑ EPE * ∆ t
k k =1
k
* df k
CVA δCVA ∂CVA * ∆x + Spread * * ∆σ + ∂ * ∆Spread * ∆x + . . . + Spread * δσ ∂x ∂Spread∂x
2
Delta hedge spread risk in proportional to discounted EPE curve.
Delta hedge potential changes in market rates (Interest rates and implied vol) in proportion to CP spread.
Potential risk of correlation of change in spread and changes in market rates.
Determines notional value of credit default swaps to buy.
Determines notional value of interest rate option need to buy to hedge interest rate risk and implied vol risk (or more general hedges needed to hedge more general market risk).
Determines sensitivity to correlation risk, which may not be possible to delta hedge. Correlation can be right or wrong way.
Effectively have transformed counterparty credit risk into market risk.
Evan Picoult, February, 2008
51
9. Summary of alternative means of hedging counterparty credit risk
Evan Picoult, February, 2008
52
CVA
Market Value CP Portfolio = Σ MVCP Portfolio (risk free)  CVACP Portfolio Two choices: – Full Market Risk Perspective:
CVA is treated as part of market risk, in the same way that the spread of a bond is treated as part of market risk. Dynamically delta hedging CVA is equivalent to transforming counterparty credit risk to market risk and hedging it like market risk. A key market variable is the implied correlation between level of rates and spread. The value of a traded CCDS enables one to impute a market implied correlation to the transaction.
– Credit Risk Perspective
CVA is treated as part of credit risk. This bifurcation of counterparty risk into a) market risk and b) credit risk is identical to what occurs under accrual accounting, where the risk of a loan portfolio is bifurcated into a) accrual interest rate risk and b) credit risk. From this perspective hedging counterparty risk is done by buying a CCDS from a third party, in the same way that the credit risk of a loan is hedged by buying a CDS or a Guarantee from a third party. US Regulators have indicated they are open to allowing a CCDS, whose reference derivative is plain vanilla, to be treated like a CDS or a Guarantee under Basel II.
Evan Picoult, February, 2008
53
This action might not be possible to undo. Are you sure you want to continue?