Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Multi-Period Credit Risk Analysis: A Macro-Scenario ApproachJUAN LICARI, SENIOR DIRECTOR
HEAD OF ECONOMIC & CONSUMER CREDIT ANALYTICS - EMEASponsored by:
22
Multi-Period Credit Risk Analysis
A Macro-Scenario Approach
• The problem we are solving
Overcome the challenge of analyzing credit risk dynamically (multi-period) and
integrate stress testing (credit risk, PPNR, provisions, mark-to-market) into a
unified management framework.
• Motivation
Federal Reserve Bank of New York - Staff Report No. 696 – Nov 2014
Supervisory Stress Tests by Beverly Hirtle & Andreas Lehnert.
“The Basel Market Risk Amendment – finalized in 1995 – contained a provision
encouraging the use of stress tests to augment the Value at Risk (VaR)
measures of computing risk-weighted assets, the denominators in various
measures of risk-based capital. VaR models consider the probability distribution
of the value of a portfolio of assets. In principle VaR models can be thought of
as the result of thousands of individual scenarios, weighted by their probability.
In practice however the distributions are not tied to real-world variables other
than the observed empirical distributions of the values of various assets.”
33
Multi-Period Credit Risk Analysis
A Macro-Scenario Approach (cont.)
Why is this relevant?
(i) embed stress testing process into the banks’ credit management practices
(forecasting, provisioning, profitability, risk-based pricing, risk concentration,
capital adequacy), (ii) dynamic risk and profitability assessment, (iii) optimal
capital management, (iv) risk appetite.
Specifics of what we do
Compute analytical solutions to multi-period credit risk management.
(1) Dynamic simulations of macro scenarios and probabilities for each path.
(2) Multi-period risk parameter simulations (leveraging stress testing models)
for conditional credit and mark-to-market parameter realizations.
(3) Analytical calculation of (intra-period and cumulative) expected and
unexpected losses, and the corresponding asset contributions.
(4) Scenario-specific analysis: embedding ST into the framework.
(5) Further applications: risk-based pricing, concentration and tail-risk analysis,
analytical reverse stress testing, dynamic optimization, integrated mark-to-
market and default risk assessment, enhanced Monte Carlo methods.
Dynamic Simulations of Macro Scenarios and Probabilities for Each Path
1
55
Forward-Looking Scenario Generation & Severity Ranking0
.05
.1.1
5.2
.25
Den
sity
-5 0 5 10 15 20Max Drop in Cumulative GDP Growth, %
Density Function
-6-4
-20
24
GD
P G
row
th,
% Q
/Q
+Q1 +Q2 +Q3 +Q4 +Q5 +Q6 +Q7 +Q8 +Q9
GDP Growth, % Q/Q: Forecasts per Quarter
Example of a Marginal Loading into Overall Scenario
Rank-Ordering Algorithm
66
Severity of Alternative Macroeconomic Scenarios
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1 2 3 4 5 6 7 8 9
Severity_0_50 Severity_0_90 Severity_0_95Severity_0_99 Severity_0_999 Severity_0_9999
GDP growth, q/q % -- alternative severity points
Dynamic Simulations of Credit Risk Parameters (Retail Credit As Leading Example)
2
88
U.S. First Mortgage Portfolio – Quarterly Vintages0
.00
5.0
1.0
15
.02
.02
5
Defa
ult R
ate
s (
#)
+Q1 +Q2 +Q3 +Q4 +Q5 +Q6 +Q7 +Q8 +Q9
Simulated values from +Q1 to +Q9 - Old/Seasoned Vintage
US First Mortgages - Vintage PD Simulations
0
.00
5.0
1.0
15
Defa
ult R
ate
s (
#)
+Q1 +Q2 +Q3 +Q4 +Q5 +Q6 +Q7 +Q8 +Q9
Simulated values from +Q1 to +Q9 - Mid-age Vintage
US First Mortgages - Vintage PD Simulations
0
.00
5.0
1.0
15
Defa
ult R
ate
s (
#)
+Q1 +Q2 +Q3 +Q4 +Q5 +Q6 +Q7 +Q8 +Q9
Simulated values from +Q1 to +Q9 - Young Vintage
US First Mortgages - Vintage PD Simulations
0
.00
1.0
02
.00
3.0
04
Defa
ult R
ate
s (
#)
+Q1 +Q2 +Q3 +Q4 +Q5 +Q6 +Q7 +Q8 +Q9
Simulated values from +Q1 to +Q9 - Future Vintage
US First Mortgages - Vintage PD Simulations
99
U.S. First Mortgage Portfolio – Quarterly Vintages (cont.)0
.00
5.0
1.0
15
.02
Block 1 Block 2 Block 3
Simulated values for +Q5 - Old/Seasoned Vintage
US First Mortgages - PD Simulations over Scenario Blocks
.00
1.0
02
.00
3.0
04
.00
5
Block 1 Block 2 Block 3
Simulated values for +Q5 - Young Vintage
US First Mortgages - PD Simulations over Scenario Blocks
0
.00
2.0
04
.00
6.0
08
Block 1 Block 2 Block 3
Simulated values for +Q5 - Mid-age Vintage
US First Mortgages - PD Simulations over Scenario Blocks
.00
01
.00
02
.00
03
.00
04
.00
05
.00
06
Block 1 Block 2 Block 3
Simulated values for +Q5 - Future Vintage
US First Mortgages - PD Simulations over Scenario Blocks
Analytical Calculation of (Intra-period and Cumulative) Expected and Unexpected Losses, Asset Contributions.
3
1111
U.S. First Mortgage Portfolio – Loss MetricsIntra-Period Exp
Loss
Intra-Period
Sigma
Intra-Period
VaR(99.9%)
Cumulative Exp
Loss
Cumulative
Sigma
Cumulative
VaR(99.9%)
+Q1 7,784.9558$ 648.5309$ 10,078.3132$ 7,784.9558$ 648.5309$ 10,078.3132$
+Q2 7,587.3311$ 850.0719$ 10,828.4443$ 15,372.2869$ 1,364.6327$ 20,248.9755$
+Q3 7,442.0702$ 999.0752$ 11,365.3219$ 22,814.3572$ 2,193.1532$ 30,983.0449$
+Q4 7,354.4725$ 1,121.8841$ 12,104.1981$ 30,168.8297$ 3,118.9971$ 42,307.3145$
+Q5 7,274.9933$ 1,228.1673$ 12,933.1039$ 37,443.8230$ 4,124.4656$ 54,375.8396$
+Q6 7,206.7848$ 1,323.8555$ 13,583.7713$ 44,650.6077$ 5,207.1490$ 66,748.3775$
+Q7 7,113.0817$ 1,407.5700$ 14,128.6149$ 51,763.6895$ 6,352.9917$ 79,727.7836$
+Q8 7,041.5606$ 1,483.2394$ 14,961.4969$ 58,805.2500$ 7,553.4249$ 93,484.9177$
+Q9 6,989.0686$ 1,559.7431$ 15,795.7796$ 65,794.3186$ 8,805.0694$ 107,915.7781$
$600
$800
$1,000
$1,200
$1,400
$1,600
$1,800
$6,000
$8,000
$10,000
$12,000
$14,000
$16,000
+Q1 +Q2 +Q3 +Q4 +Q5 +Q6 +Q7 +Q8 +Q9
Une
xpec
ted
Loss
(Sig
ma)
Expe
cted
Los
ses
and
VaR(
99.9
%)
Intra-Period Expected Losses, σ and VaR(99.9%) (Millions of USD)
Intra-Period Exp Loss
Intra-Period VaR(99.9%)
Intra-Period Sigma
1212
U.S. First Mortgage Portfolio – Loss Metrics (cont.)
1313
U.S. First Mortgage Portfolio – Loss Metrics (cont.)
1414
U.S. First Mortgages – Tail-Risk Contributions
Scenario-Specific Analysis: Embedding ST Into Dynamic Portfolio Analysis
4
1616
U.S. First Mortgages – Intra-Period Scenario Losses
$5,000
$6,000
$7,000
$8,000
$9,000
$10,000
$11,000
$12,000
$13,000
$14,000
+Q1 +Q2 +Q3 +Q4 +Q5 +Q6 +Q7 +Q8 +Q9
Scenario-Specific Intra-Period Losses (Millions of USD)
CCAR Baseline
ECCA's s1
ECCA's Baseline
ECCA's s2
ECCA's s5
CCAR Adverse
ECCA's s3
ECCA's s6
CCAR Severely Adverse
ECCA's s4
1717
U.S. First Mortgages – Cumulative Scenario Losses
CCAR bl
CCAR bl
CCAR Adv
CCAR Adv
CCAR Sev Adv
CCAR Sev Adv
CCAR bl
CCAR AdvCCAR Sev Adv
CCAR bl
CCAR AdvCCAR Sev Adv
1818
U.S. First Mortgages – Cumulative Scenario Losses (cont.)1
8
0.277818
0.947129
0.998805
0.231698
0.138336
0.412587
0.986158 0.998419
0.466025
0.958049
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
VaR
_C
CA
R_b
l
VaR
_C
CA
R_a
VaR
_C
CA
R_s
a
VaR
_EC
CA
_bl
VaR
_EC
CA
_s1
VaR
_EC
CA
_s2
VaR
_EC
CA
_s3
VaR
_EC
CA
_s4
VaR
_EC
CA
_s5
VaR
_EC
CA
_s6
Prob(Cumulative L
1919
U.S. First Mortgages – Cumulative Scenario Losses (cont.)
Practical Applications: RAROC, Optimization, Optimal Importance Sampling, and Reverse Stress Testing
5
2121
Analytical calculation of the portfolio volatility up to the new deal’s
maturity allows us to instantaneously price.
The Sharpe ratio of the new portfolio with the new deal (or deals)
should be larger than without:
Risk Adjusted Pricing (RAROC)
𝜇∗ − 𝑟
𝜎 𝐿∗>𝜇 − 𝑟
𝜎 𝐿BUY
2222
What is the portfolio composition 𝑛𝑖 that minimises the portfolio loss volatility given a level of expected loss (and hence return) 𝐸𝐿 = 𝐿?
Using the Lagrange multipliers:
Thus:
- Extend the current framework to study DYNAMIC OPTIMISATION (infinite horizon)Recursive Dynamic Programming (Bellman Equations) and the study of the optimal
solutions to the underlying stochastic difference equations
Portfolio Optimisation
Λ 𝑛𝑖 , 𝜆 = 𝜎 𝐿; 𝑛𝑖 + 𝜆 𝐸𝐿 𝑛𝑖 − 𝐿
𝑛𝑖𝐶𝑖 + 𝜆𝐸𝐿𝑖∗ = 0
𝑖=1
𝑁
𝑛𝑖𝐸𝐿𝑖∗ − 𝐿 = 0
0
1
2
3
4
5
6
7
8
9
10
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Frontiers of Optimal PortfoliosOriginal Allocations, Constrained (no short-selling) and Unconstrained Frontiers
ORIGINAL ALLOCATIONS
UNCOSTRAINED OPTIMA (Short-Selling)
CONSTRAINED OPTIMA (No Short-selling)
AUTO LOAN Portfolio Only
MORTGAGE Portfolio Only
CREDIT CARD Portfolio Only
2323
Instead of drawing random numbers with unconditional probability from conditional
distribution 𝑤𝑧 draw from conditional 𝑤𝑧 𝛼 . After the simulation reweigh the loss
distribution using weight =𝑤𝑧
𝑤𝑧 𝛼
Optimal Importance Sampling
-6 -4 -2 0 2 4 6
0.0
00
.01
0.0
20
.03
0.0
4
Z
Pro
ba
bility
-6 -4 -2 0 2 4 6
0.0
00
.02
0.0
40
.06
0.0
8
Z
Pro
ba
bility
-300 -200 -100 0 100 200 300 400
0.0
00
0.0
01
0.0
02
0.0
03
0.0
04
0.0
05
Total Loss in period 4
loss
pro
ba
bility
2424
Analytical Reverse Stress Testing
Mark-to-Market Risk ExamplesInterest Rates (Gov Bond Yields) and Corporate Credit Spreads
6
2626
Interest Rate Modeling
Impact on government bond yields
05
10
15
20
Yie
lds, %
3m 6m 1y 2y 3y 5y 7y 10y 20y 30y
Box-Plots Across Maturities at +Q9
Government Bond Yield Curves
2727
Interest Rate Modeling (cont.)
Impact on government bond yields (cont.)
0.2
.4.6
Density
0 2 4 6 8 10Term Premium (10y vs. 3m Spread, %)
Distribution for +Q9
0 2 4 6 8 10Yield Curve Slope (10y vs. 3m Spread, %)
Box-Plot for +Q9
02
46
810
Yie
ld C
urv
e S
lope
0 5000 10000 15000 20000 25000Simulation id (1 for Good , 25000 for Stressed)
+Q9 Values over Simulations
02
46
810
Yie
ld C
urv
e S
lope
+Q1 +Q2 +Q3 +Q4 +Q9 +Q6 +Q7 +Q8 +Q9
Box-Plots, all +Qs
Analysis Over Quarters and Simulations
Yield Curve Slope (10y vs. 3m Spread, %)
2828
Credit Spread ModelingImpact on credit spreads for financials - over maturities and rating classes
0.5
11.5
2
Spre
ad, %
3m 1y 3y 5y 7y 10y 20y 30y
Aaa
05
10
15
Spre
ad, %
3m 1y 3y 5y 7y 10y 20y 30y
A
05
10
15
20
Spre
ad, %
3m 1y 3y 5y 7y 10y 20y 30y
Bbb
05
10
15
20
25
Spre
ad, %
3m 1y 3y 5y 7y 10y 20y 30y
B
5 Quarters out of Sample (+Q5) - Across Rating Classes
Corporate Spread Curves - Financials
www.economy.com
United States
121 North Walnut Street
Suite 500
West Chester PA 19380
+1.610.235.5299
Australia
Level 10
1 O'Connell Street
Sydney, NSW, 2000
Australia
+61.2.9270.8111
United Kingdom
One Canada Square
Canary Wharf
London E14 5FA
+44.20.7772.5454
Prague
Washingtonova 17
110 00 Prague 1
Czech Republic
+420.22.422.2929
3030
© 2015, Moody’s Analytics, Inc. and/or its licensors and affiliates (collectively, “MOODY’S”). All rights reserved. ALL INFORMATION CONTAINED HEREIN IS
PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR OTHERWISE REPRODUCED, REPACKAGED, FURTHER
TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN
WHOLE OR IN PART, IN ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER, BY ANY PERSON WITHOUT MOODY’S PRIOR WRITTEN CONSENT.
All information contained herein is obtained by MOODY’S from sources believed by it to be accurate and reliable. Because of the possibility of human or
mechanical error as well as other factors, however, all information contained herein is provided “AS IS” without warranty of any kind. Under no circumstances shall
MOODY’S have any liability to any person or entity for (a) any loss or damage in whole or in part caused by, resulting from, or relating to, any error (negligent or
otherwise) or other circumstance or contingency within or outside the control of MOODY’S or any of its directors, officers, employees or agents in connection with
the procurement, collection, compilation, analysis, interpretation, communication, publication or delivery of any such information, or (b) any direct, indirect, special,
consequential, compensatory or incidental damages whatsoever (including without limitation, lost profits), even if MOODY’S is advised in advance of the possibility
of such damages, resulting from the use of or inability to use, any such information. The ratings, financial reporting analysis, projections, and other observations, if
any, constituting part of the information contained herein are, and must be construed solely as, statements of opinion and not statements of fact or
recommendations to purchase, sell or hold any securities. NO WARRANTY, EXPRESS OR IMPLIED, AS TO THE ACCURACY, TIMELINESS, COMPLETENESS,
MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR OTHER OPINION OR INFORMATION IS GIVEN OR MADE
BY MOODY’S IN ANY FORM OR MANNER WHATSOEVER. Each rating or other opinion must be weighed solely as one factor in any investment decision made
by or on behalf of any user of the information contained herein, and each such user must accordingly make its own study and evaluation of each security and of
each issuer and guarantor of, and each provider of credit support for, each security that it may consider purchasing, holding, or selling.