Managing Credit Risk across a Diverse Corporate Portfolio Tarun Dara, Mubadala GE Capital Confidential
Jul 08, 2015
Managing Credit Risk across a Diverse Corporate Portfolio
Tarun Dara, Mubadala GE Capital
Confidential
“It is not the strongest or the most intelligent who will survive but those who can best manage change” - Charles Darwin
Confidential 2
5 C’s of Credit Character Capacity Capital Collateral Conditions Profitability Yield is greater than threshold set
Traditional View: Individual Transaction
Confidential 3
• Probability of Default Within 1 year
• Default = Bankruptcy,
90 Days Past Due,
Restructuring to Avoid Default
• Economic Losses Post Default
• Includes All Costs, Timing of Recoveries
• Function Of Collateral, Seniority
• Expected Size of Exposure at Default
• Need to model for Revolvers & for other unfunded commitments
• “Cost Of Doing Business”
• Given Predictable, not “Risk” per se
• Typically reflected in Pricing
Probability of Default
Loss Given
Default
Exposure At Default
Expected Losses
= X X
Credit Risk Quantified : Individual Transaction
Confidential 4
75
380
75 100
225
150 145
60 60 70 100
60 60
265
175
0
50
100
150
200
250
300
350
400
0.23% 0.27% 0.30% 0.31% 0.34% 0.36% 0.45% 0.47% 0.50% 0.54% 0.60% 0.68% 0.90% 0.91% 1.03%
EAD
($
mm
)
EL %
50
350
635
245 320
225 175
0
100
200
300
400
500
600
700
0.59% 0.89% 1.34% 2.01% 3.03% 4.55% 6.84%
EAD
($
mm
)
PD %
75
1060
705
160
0
200
400
600
800
1000
1200
0%-15% 15%-30% 30%-45% 45%-65%
EAD
($
mm
)
LGD %
Credit Risk Measures for a Diverse Portfolio
PD Distribution LGD Distribution
EL Distribution
Confidential 5
At deal level we understand the “risk characteristics”
Confidential
What should be the portfolio composition ?
How do we maximize portfolio return ?
6
Economic Capital is Provided to Cover Extreme or “Tail Risk” Events
EL
Probability of a “Tail Risk” Event is a function of Unexpected Loss (Volatility)
Expected Loss Vs. Unexpected Loss
Confidential 7
Low
High
Pro
bab
ility
of
Loss
Very Small Size of Loss Very Large
1
EL
1
“Tail Risk” loss event 99.9% Confidence Level
3
UL
2
Average Loss (Expected Loss or “EL” )
Volatility (Unexpected Loss or “UL”)
Extreme Loss or “Tail Risk”
2
3
Portfolio Risk Measures
Confidence level scaled to target rating level
Confidential 8
Ecap (Economic Capital) is the amount of capital a business or product require, for a given amount of risk, to achieve a specified level of confidence that an extreme loss will not exceed the capital
Measured as the potential loss in excess of EL over one year time period at a specified confidence level or criterion
Represents comprehensive risk measurement which can be thought of as “unit of risk”
Economic Capital
Confidential 9
Business risk Structural
ALM Equity
risk Operational
risk Credit
risk
Inter-risk correlations
Overall EC
Comprehensive Ecap Framework
- Losses resulting from inadequate or failed internal processes, people and systems or from external events
- Change in Economic Value of Equity under interest rate and FX scenarios
- Change in value of Equity investments under market scenarios
- Losses resulting from volume and margin decline
- Change in value of credit positions due to default and credit migrations - Most substantial risk type
Confidential 10
Two approaches to measure Credit Risk Ecap In-house formula based model Moody’s Risk Frontier
Credit Risk Ecap: Measurement
Confidential 11
Critical Assumptions: Ratings S&P, KMV,
FICO etc.
Loss History (Recoveries, Collateral
etc.) Outstanding's at default
Probability of Default PD
Loss Given Default LGD
Exposure at Default EAD
Expected Loss EL
x x =
Unexpected Loss ( UL)
(Standard deviation in Expected Loss)
Credit Risk Economic Capital
Credit Rating Capital Multiplier
Desired Credit Rating
Diversification Credit Default Correlations
Historical Correlation of the Portfolio
• Cycle-Neutral
• Benchmark Correlation / Diversification Assumption
• Benchmark Capital Multiplier
• 1 Year Time Horizon
Model Formula
Confidential 12
• A simulation approach to calculate the capital required to maintain a certain level of risk in a credit portfolio
• Allocates capital to individual exposures in the portfolio
• Uses multiple factors based on the global economy, region, sector, industry and country
• Correlation between any two firms’ returns can be explained by the firms’ relationship to a set of common factors
• The greater the link of two individual firms to common economic factors, the greater the likelihood that their fortunes will rise and fall together
Moody’s Risk Frontier
Confidential 13
Systematic
Idiosyncratic
Idiosyncratic
Systematic
Industry, Country and Sales Size determine how much risk is systematic (undiversifiable) vs. idiosyncratic (diversifiable)
Blue Chip MNC has very low stand-alone risk, but that risk is mainly associated to factors that affect the entire economic system and, therefore, all other obligors in a portfolio (i.e. it cannot be diversified away).
Best Pizza, LLC is very risky, but its risk is mainly due to factors that are specific to that company (e.g. neighborhood, health of owner/manager, competition on the street, local economy) and, therefore, not correlated with other obligors in a portfolio.
Low R2 High R2
Modeling Co-relation
Best Pizza LLC (PD : 0.1029) Blue Chip MNC (PD : 0.0003)
Confidential 14
Higher levels of LGDs are associated with increasing capital charges, even though the PDs decrease enough to keep the EL at the same level
Given the same EL, it’s always better to have lower LGD
Capital Sensitivity to LGD
Confidential 15
Australia, $220mm
USA $440mm
UK , $190mm Germany, $ 120mm
Japan, $50mm
UAE, $160 mm
Egypt, $310 mm
KSA, $390mm
Turkey, $120mm
Country, EAD (This is sample data set created for this discussion)
Diverse Corporate Portfolio
Confidential 16
Australia Germany
Japan
UAE
KSA
Turkey
Egypt
UK
USA
5.0%
7.0%
9.0%
11.0%
13.0%
15.0%
17.0%
19.0%
21.0%
0.25% 0.35% 0.45% 0.55% 0.65% 0.75% 0.85% 0.95%
Ecap
(%)
EL(%)
Ecap Requirements: Country Level
KSA has same level of EL as UK but significantly higher Ecap
Distinguish attractiveness of countries at same level of EL
Avg
. 0
.52
%
Avg. 13.0%
UAE & Turkey attract more capital than Australia despite similar EL levels
Germany, Japan look very attractive
Confidential 17
Bubble Size is based on the EAD of Sample Country
Ecap Requirements: Product Level
Asset Finance
Corporate Loan
Project Finance
Real Estate Working Capital
7.0%
8.0%
9.0%
10.0%
11.0%
12.0%
13.0%
14.0%
15.0%
16.0%
17.0%
0.20% 0.30% 0.40% 0.50% 0.60% 0.70% 0.80%
Ecap
(%)
EL(%)
Asset Finance EL level is close to highest but has the lowest Ecap
Working Capital has higher EL than Corp Loans but lower Ecap
Attractiveness at product level….
Avg
. 0
.52
%
Avg. 13.0%
Confidential 18
Bubble Size is based on the EAD of Sample Product
I
K
L
M
N
4.0%
6.0%
8.0%
10.0%
12.0%
14.0%
16.0%
18.0%
20.0%
22.0%
24.0%
0.2% 0.3% 0.4% 0.5% 0.6% 0.7% 0.8% 0.9% 1.0% 1.1%
Ecap
(%)
EL(%)
Avg
. 0.5
2%
Avg. 13.0%
Ecap Requirements: Deal Level
Deals in bottom right have lower capital requirements and in top left have higher capital requirements Confidential 19
Bubble Size is based on the Sample Deal EAD
D
G
I
K
M O
P
Q
R
T
V
W
Z
3.0%
4.0%
5.0%
6.0%
7.0%
8.0%
9.0%
3.0% 8.0% 13.0% 18.0% 23.0% 28.0%
Yie
ld
Ecap(%)
Ecap vis-à-vis Yields
Avg
. 13
%
Avg. 6.45%
Deals in top left are most profitable and the mix has changed Confidential 20
Bubble Size is based on the Sample Deal EAD
B
4.0%
8.0%
12.0%
16.0%
20.0%
24.0%
28.0%
4.0% 8.0% 12.0% 16.0% 20.0% 24.0% 28.0%
RA
RO
C
Ecap(%)
Ecap vis-à-vis Risk Adjusted Returns
Lim
it 2
0%
Hurdle 20%
Deals in top left quadrant meet the hurdle rate and Ecap % limit Confidential
21 Bubble Size is based on the Sample Deal EAD
Concluding Remarks
This framework is meant to supplement the underwriting of each
credit. Continue to maintain high standards of underwriting A Deal looking attractive on its own may not enhance portfolio
returns if it is carrying similar “risk characteristics” Diversification of portfolio is key for return maximization
Mechanism to proactively manage the portfolio
Extend this to perform stress testing which has taken a very
important role from regulatory standpoint
Confidential 22
Thank You
Confidential 23
Mubadala GE Capital Mubadala GE Capital PJSC is a specialized commercial finance company providing structured financing solutions to businesses across MENAT. Headquartered in Abu Dhabi and owned by Mubadala Development Company & GE Capital.
Tarun Dara Tarun is currently leading credit risk and portfolio analytics at Mubadala GE Capital. Over last 13 years he has worked in area of underwriting, corporate finance, portfolio analytics and financial risk consulting spread across India, Asia Pacific and Middle East region. His current focus is on building a robust portfolio and capital management framework, governance of risk models in use across diverse asset classes and portfolio quality reviews. Tarun holds an MBA in finance and strategy from IIT, Delhi ; MS (Hons) Economics from BITS, Pilani in India and has graduated from a financial management program with GE.
Confidential 24