Stress & Reverse Stress from a Macro Viewpoint€¦ · 7 Stress Testing: 4- Modelling Methodologies...
Transcript of Stress & Reverse Stress from a Macro Viewpoint€¦ · 7 Stress Testing: 4- Modelling Methodologies...
Stress & Reverse Stress from a Macro Viewpoint
Dr. Juan M. Licari
Senior Director
Head of Economic & Credit Analytics - EMEA
2
Stress & Reverse Stress Testing from a Macro Viewpoint
Key Stress Testing Challenges:
1- Dynamic vs. Static Approach to Stress Testing,
2- Partial vs. General Equilibrium,
3- Top-down vs. Bottom-up,4- Modelling Methodologies: Stress Testing vs.
Forecasting/Scoring,
5- Quantitative Reverse Stress Testing,
6- Looking beyond Capital & Solvency.
Historic and predicted default rates (baseline),
% of balance at origination
Consolidated portfolio, vintages over time
Performance of Future Loans
Forecasted Performance of Existing LoansPerformance History
3
Stress Testing: 1- Dynamic vs. Static Approach
4
Stress Testing: 2- Partial vs. General EquilibriumExample of top-down stress testing approach, Bank of England’s RAMSI model
5
Stress Testing: 2- Partial vs. General Equilibrium
» Interest rates
» Unemployment rates
» Income growth
» Profits (National Accounts)
» Share market
Small Business Loans
» Interest rates
» Unemployment rate
» Commodity/oil prices
» Price index for used cars
Auto-Equipment Loan/Lease
Illustrative
» Mortgage rate difference from origination
» Unemployment rate
» Employment growth
» Income growth
» House price growth
» Home equity
RMBS
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
2009
M06
2009
M10
2010
M02
2010
M06
2010
M10
2011
M02
2011
M06
2011
M10
2012
M02
2012
M06
2012
M10
2013
M02
2013
M06
2013
M10
Baseline
S3
S4
DD
0
10
20
30
40
50
60
70
80
90
10020
09M
0620
09M
1120
10M
0420
10M
0920
11M
0220
11M
0720
11M
1220
12M
0520
12M
1020
13M
0320
13M
0820
14M
0120
14M
0620
14M
1120
15M
0420
15M
09
Baseline
S3
S4
DD
PD term-structure
LGD curves
Examples of collateral type for RMBS/ABS deals
6
Stress Testing: 3- Top-down vs. Bottom-up
Issue: Loan level model can miss correlations and feedback effects
» Individual performance depends on other loans
» Difficult to model individuals within a system
Risk models could miss the forest for the trees
– Why not model the forest, model the trees and then make sure
the tree model agrees with forest projections?
≠
7
Stress Testing: 4- Modelling Methodologies0
.2.4
.6.8
1
Tra
nsitio
n %
2000m1 2002m1 2004m1 2006m1 2008m1 2010m1 2012m1Month of Transition
Binary_Probit_Regression O_1_Median_Variable
0.2
.4.6
.81
Tra
nsitio
n %
2000m1 2002m1 2004m1 2006m1 2008m1 2010m1 2012m1Month of Transition
Binary_Probit_Regression O_1_Median_Variable
Binary (Probit) Model Downgrade0
.1.2
.3.4
Tra
nsit
ion
%
2000m1 2002m1 2004m1 2006m1 2008m1 2010m1 2012m1 2014m1 2016m1Month of Transition
Actuals Baseline
FSA Scenario4
Custom
0.0
2.0
4.0
6.0
8T
ran
sitio
n %
2000m1 2002m1 2004m1 2006m1 2008m1 2010m1 2012m1 2014m1 2016m1Month of Transition
Actuals Baseline
FSA Scenario4
Custom
CaaC to DefaultBaa to A
Binary (Probit) Model Upgrade
-60
-40
-20
0
20
40
60
20
01
M9
20
02
M4
20
02
M1
1
20
03
M6
20
04
M1
20
04
M8
20
05
M3
20
05
M1
0
20
06
M5
20
06
M1
2
20
07
M7
20
08
M2
20
08
M9
20
09
M4
20
09
M1
1
20
10
M6
20
11
M1
20
11
M8
20
12
M3
20
12
M1
0
20
13
M5
20
13
M1
2
20
14
M7
20
15
M2
20
15
M9
20
16
M4
Equity Forecasts, Annual Returns
tyy_ukxindex_bl
tyy_ukxindex_fsa
tyy_ukxindex_s4
-60
-40
-20
0
20
40
60
20
01
M9
20
02
M4
20
02
M1
1
20
03
M6
20
04
M1
20
04
M8
20
05
M3
20
05
M1
0
20
06
M5
20
06
M1
2
20
07
M7
20
08
M2
20
08
M9
20
09
M4
20
09
M1
1
20
10
M6
20
11
M1
20
11
M8
20
12
M3
20
12
M1
0
20
13
M5
20
13
M1
2
20
14
M7
20
15
M2
20
15
M9
20
16
M4
Equity Forecasts, Annual Returns
tyy_spxindex_bl
tyy_spxindex_fsa
tyy_spxindex_s4
-60
-40
-20
0
20
40
60
20
01
M8
20
02
M3
20
02
M1
0
20
03
M5
20
03
M1
2
20
04
M7
20
05
M2
20
05
M9
20
06
M4
20
06
M1
1
20
07
M6
20
08
M1
20
08
M8
20
09
M3
20
09
M1
0
20
10
M5
20
10
M1
2
20
11
M7
20
12
M2
20
12
M9
20
13
M4
20
13
M1
1
20
14
M6
20
15
M1
20
15
M8
20
16
M3
Equity Forecasts, Annual Returns
tyy_mxwoindex_bl
tyy_mxwoindex_fsa
tyy_mxwoindex_s4
8
Equity Indexes, annual growth rateHistory and forecasts
World US
UK
Stress Testing: 4- Modelling Methodologies
9
-40
-20
0
20
40
60
80
100
120
20
02
M1
20
02
M7
20
03
M1
20
03
M7
20
04
M1
20
04
M7
20
05
M1
20
05
M7
20
06
M1
20
06
M7
20
07
M1
20
07
M7
20
08
M1
20
08
M7
20
09
M1
20
09
M7
20
10
M1
20
10
M7
20
11
M1
20
11
M7
20
12
M1
20
12
M7
20
13
M1
20
13
M7
20
14
M1
20
14
M7
20
15
M1
20
15
M7
20
16
M1
20
16
M7
20
17
M1
20
17
M7
20
18
M1
Corporate Spread, Aaa Financials, 03m
f_aaa_f3m_bl
f_aaa_f3m_fsa
f_aaa_f3m_s4
0
50
100
150
200
250
300
20
02
M1
20
02
M7
20
03
M1
20
03
M7
20
04
M1
20
04
M7
20
05
M1
20
05
M7
20
06
M1
20
06
M7
20
07
M1
20
07
M7
20
08
M1
20
08
M7
20
09
M1
20
09
M7
20
10
M1
20
10
M7
20
11
M1
20
11
M7
20
12
M1
20
12
M7
20
13
M1
20
13
M7
20
14
M1
20
14
M7
20
15
M1
20
15
M7
20
16
M1
20
16
M7
20
17
M1
20
17
M7
20
18
M1
Corporate Spread, Aaa Financials, 10y
f_aaa_f10y_bl
f_aaa_f10y_fsa
f_aaa_f10y_s4
-100
-50
0
50
100
150
200
20
02
M1
20
02
M8
20
03
M3
20
03
M1
0
20
04
M5
20
04
M1
2
20
05
M7
20
06
M2
20
06
M9
20
07
M4
20
07
M1
1
20
08
M6
20
09
M1
20
09
M8
20
10
M3
20
10
M1
0
20
11
M5
20
11
M1
2
20
12
M7
20
13
M2
20
13
M9
20
14
M4
20
14
M1
1
20
15
M6
20
16
M1
20
16
M8
20
17
M3
20
17
M1
0
20
18
M5
Corporate Spread, Aaa Non-Financials, 03m
f_aaa_nf3m_bl
f_aaa_nf3m_fsa
f_aaa_nf3m_s4
-50
0
50
100
150
200
250
300
20
02
M1
20
02
M7
20
03
M1
20
03
M7
20
04
M1
20
04
M7
20
05
M1
20
05
M7
20
06
M1
20
06
M7
20
07
M1
20
07
M7
20
08
M1
20
08
M7
20
09
M1
20
09
M7
20
10
M1
20
10
M7
20
11
M1
20
11
M7
20
12
M1
20
12
M7
20
13
M1
20
13
M7
20
14
M1
20
14
M7
20
15
M1
20
15
M7
20
16
M1
20
16
M7
20
17
M1
20
17
M7
20
18
M1
Corporate Spread, Aaa Non-Financials, 10y
f_aaa_nf10y_bl
f_aaa_nf10y_fsa
f_aaa_nf10y_s4
Stress Testing: 4- Modelling Methodologies
10
Stress Testing: 5- Reverse Stress Testing
Portfolio Expected Values – Defining Severity
Scenario 2
5, 10, 100 bps
Baseline
5, 10, 100 bps
Incr. SR
11
Stress Testing: 5- Reverse Stress Testing
12
Analyzing Factors’ Realizations Across Tail Events
Stress Testing: 5- Reverse Stress Testing
13
-50
5g
dp
gro
wth
-3-2
-10
1fa
cto
r 1
2000q1 2003q1 2006q1 2009q1 2012q1
factor 1 gdp growth
2000Q1-2011Q4: Factor 1 vs GDP growth.
-3
-2
-1
0
1
2
3
4
5
2006 2007 2008 2009 2010 (E) 2011 (F) 2012 (F) 2013 (F) 2014 (F)
Simulations of Inflation Rate, History & Forecasts, Euro-
Zone Level
0
50
100
150
200
250
300
less
th
an
-3
-3 t
o -
1.5
-1.5
to
0
0 t
o 1
.5
1.5
to
3
3 t
o 4
.5
4.5
to
6
6 t
o 7
.5
7.5
to
9
9 t
o 1
0.5
10
.5 t
o 1
2
ov
er
12
Historic
2012Q1
2014Q1
2016Q1
Simulations of Real GDP Growth, Emerging Markets
Standard Alternative Scenarios
Macro Factor Analysis
Linking Factors’ Realizations to Macro Scenarios
Stress Testing: 5- Reverse Stress Testing
14
LLLLRRRR tttteeeesssstttt:::: iiiinnnnddddeeeeppppeeeennnnddddeeeennnntttt vvvvssss.... ssssaaaattttuuuurrrraaaatttteeeedddd:::: cccchhhhiiii2222((((222211110000)))) ==== 1111444444443333....77770000 PPPPrrrroooobbbb>>>>cccchhhhiiii2222 ==== 0000....0000000000000000 FFFFaaaaccccttttoooorrrr22221111 ----0000....00005555777722228888 .... ----0000....0000000033330000 1111....0000000000000000 FFFFaaaaccccttttoooorrrr22220000 ----0000....00004444444400002222 0000....00001111333322226666 ----0000....0000000022223333 1111....0000000033330000 FFFFaaaaccccttttoooorrrr11119999 ----0000....00003333111144449999 0000....00001111222255553333 ----0000....0000000011116666 1111....0000000055553333 FFFFaaaaccccttttoooorrrr11118888 ----0000....00002222111166669999 0000....00000000999988880000 ----0000....0000000011111111 1111....0000000066669999 FFFFaaaaccccttttoooorrrr11117777 ----0000....00001111555544442222 0000....00000000666622227777 ----0000....0000000000008888 1111....0000000088881111 FFFFaaaaccccttttoooorrrr11116666 ----0000....00001111111144445555 0000....00000000333399997777 ----0000....0000000000006666 1111....0000000088889999 FFFFaaaaccccttttoooorrrr11115555 ----0000....00000000666622222222 0000....00000000555522223333 ----0000....0000000000003333 1111....0000000099995555 FFFFaaaaccccttttoooorrrr11114444 0000....00000000222222222222 0000....00000000888844444444 0000....0000000000001111 1111....0000000099998888 FFFFaaaaccccttttoooorrrr11113333 0000....00002222111144449999 0000....00001111999922227777 0000....0000000011111111 1111....0000000099997777 FFFFaaaaccccttttoooorrrr11112222 0000....00003333555544448888 0000....00001111333399999999 0000....0000000011118888 1111....0000000088885555 FFFFaaaaccccttttoooorrrr11111111 0000....00006666222277772222 0000....00002222777722224444 0000....0000000033333333 1111....0000000066667777 FFFFaaaaccccttttoooorrrr11110000 0000....11114444555599992222 0000....00008888333322220000 0000....0000000077776666 1111....0000000033334444 FFFFaaaaccccttttoooorrrr9999 0000....22221111222233339999 0000....00006666666644447777 0000....0000111111111111 0000....9999999955558888 FFFFaaaaccccttttoooorrrr8888 0000....22228888333366668888 0000....00007777111122229999 0000....0000111144448888 0000....9999888844447777 FFFFaaaaccccttttoooorrrr7777 0000....44443333777788889999 0000....11115555444422221111 0000....0000222222228888 0000....9999777700000000 FFFFaaaaccccttttoooorrrr6666 0000....66661111777799990000 0000....11118888000000001111 0000....0000333322222222 0000....9999444477771111 FFFFaaaaccccttttoooorrrr5555 0000....99996666444499993333 0000....33334444777700003333 0000....0000555500003333 0000....9999111144449999 FFFFaaaaccccttttoooorrrr4444 1111....33339999111166668888 0000....44442222666677775555 0000....0000777722225555 0000....8888666644446666 FFFFaaaaccccttttoooorrrr3333 1111....77771111999911118888 0000....33332222777755550000 0000....0000888899996666 0000....7777999922221111 FFFFaaaaccccttttoooorrrr2222 3333....88889999444477772222 2222....11117777555555554444 0000....2222000033330000 0000....7777000022225555 FFFFaaaaccccttttoooorrrr1111 9999....55558888333311111111 5555....66668888888833339999 0000....4444999999995555 0000....4444999999995555 FFFFaaaaccccttttoooorrrr EEEEiiiiggggeeeennnnvvvvaaaalllluuuueeee DDDDiiiiffffffffeeeerrrreeeennnncccceeee PPPPrrrrooooppppoooorrrrttttiiiioooonnnn CCCCuuuummmmuuuullllaaaattttiiiivvvveeee
RRRRoooottttaaaattttiiiioooonnnn:::: ((((uuuunnnnrrrroooottttaaaatttteeeedddd)))) NNNNuuuummmmbbbbeeeerrrr ooooffff ppppaaaarrrraaaammmmssss ==== 222200003333 MMMMeeeetttthhhhoooodddd:::: pppprrrriiiinnnncccciiiippppaaaallll ffffaaaaccccttttoooorrrrssss RRRReeeettttaaaaiiiinnnneeeedddd ffffaaaaccccttttoooorrrrssss ==== 11114444FFFFaaaaccccttttoooorrrr aaaannnnaaaallllyyyyssssiiiissss////ccccoooorrrrrrrreeeellllaaaattttiiiioooonnnn NNNNuuuummmmbbbbeeeerrrr ooooffff oooobbbbssss ==== 44448888
UK Factor Analysis – Top 5 Factors
UK factor analysis – 21 macroeconomic series
Stress Testing: 5- Reverse Stress TestingLinking Factors’ Realizations to Macro Scenarios
15
-10
-50
5g
dp
gro
wth
-3-2
-10
1fa
cto
r 1
2000q1 2003q1 2006q1 2009q1 2012q1
factor 1 gdp growth
UK, 2000-2012: Factor 1 vs GDP growth
-2-1
01
2e
mp
loym
en
t g
row
th
-2-1
01
2fa
cto
r 2
2000q1 2003q1 2006q1 2009q1 2012q1
factor 2 employment growth
UK, 2000-2012: Factor 2 vs Growth in Employment
Stress Testing: 5- Reverse Stress TestingLinking Factors’ Realizations to Macro Scenarios, UK Top Factors
-6-4
-20
2in
fla
tio
n (
-)
-3-2
-10
12
facto
r 3
2000q1 2003q1 2006q1 2009q1 2012q1
factor 3 inflation (-), lagged
UK, 2000-2012: Factor 3 vs Inflation
-6-4
-20
mo
ne
tary
po
licy (
-)
-3-2
-10
12
facto
r 3
2000q1 2003q1 2006q1 2009q1 2012q1
factor 3 monetary policy (-)
UK, 2000-2012: Factor 3 vs Monetary Policy rate
-12
-10
-8-6
-4-2
M1
gro
wth
-2-1
01
2fa
cto
r 4
2000q1 2003q1 2006q1 2009q1 2012q1
factor 4 M1 growth
UK, 2000-2012: Factor 4 vs M1 (growth rate)
-30
-20
-10
01
02
0H
om
e P
rice
gro
wth
-10
12
3fa
cto
r 5
2000q1 2003q1 2006q1 2009q1 2012q1
factor 5 Home Price Index growth (lagged)
UK, 2000-2012: Factor 5 vs Home Price growth (lagged)
16
Stress Testing: 6- Looking beyond Capital & Solvency
Bank A Bank B
Bank D Bank C
Bank A Bank B
Bank D Bank C
Bank B1 Bank B2 Bank B3 Bank B4
Bank A1 Bank A2
Bank A3 Bank A4
Bank B5 Bank B6 Bank B7 Bank B8
Case 1: Incomplete Markets Case 2: Complete Markets
Case 3: A Hybrid Market Structure
17
(1) If a small institution fails, other peripheral banks are likely to suffer asset-liability mismatches. But the systemic risk involved is not as high. Large institutions should be able to absorb the original shock and stop the domino effects once it enters the core banking area.
(2) If the shock is to one of the large financial institutions, there is a secondary effect through the overall health of the financial economy that could put all institutions under tremendous pressure.
Though the core banking sub-sector is complete (bi-directional flows), the fact that it is highly concentrated poses severe challenges to the banking sector as a whole. It is not so much a direct domino effect between, say, Bank A1 and Bank A2, but an indirect
contagion risk: (2.a) Bank A1 affects the overall economy (as it is a big financial institution whose failure can put pressure on the local financial and labor markets), (2.b) the economy affects Bank A2 as an external shock, (2.c) Bank A2’s failure worsens the economic picture even more, (2.d) the economy affects Bank A3, etc.
.04
.045
.05
.055
.06
130 140 150 160 170qtime
ff_pl_pct_ast_tot_bl ff_pl_pct_ast_tot_s2
ff_pl_pct_ast_tot_s3 ff_pl_pct_ast_tot_s4
ff_pl_pct_ast_tot_s5 ff_pl_pct_ast_tot_s6
P&L Rate, Total Assets
2010 2011 2012 2013 2014
.015
.02
.025
.03
130 140 150 160 170qtime
ff_pl_pct_liab_tot_bl ff_pl_pct_liab_tot_s2
ff_pl_pct_liab_tot_s3 ff_pl_pct_liab_tot_s4
ff_pl_pct_liab_tot_s5 ff_pl_pct_liab_tot_s6
P&L Rate, Total Liabilities
2010 2011 2012 2013 2014
Stress Testing: 6- Looking beyond Capital & Solvency
© 2013 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 Fuentes 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, resultingfrom, 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.