Henry Albulescu Managing Director

1
Introduction to SPs CDO Methodology

• Henry AlbulescuManaging Director
• 1-212-438-2382
• henry_albulescu_at_sandp.com

2
3 Steps To CDO Structure

• Collateral Quality and expected Default Rates
• Recoveries on Defaults Capital Structure
• Legal and Structural Analyses

3
Step 1 - Asset Pool Default Rates

• Pool Default Rates Driven by- Asset type (
  Corporate vs. ABS)- Rating of each asset-
  Exposure to each obligor- Industry of each
  obligor- Exposure period (maturity)
• SPs CDO EVALUATOR gives Gross Pool Default
  rates- Model uses Monte Carlo simulation to
  calculate the expected default rate at each
  rating level.
• - Example AAA 40
• AA 35
• A 30
• For a tranche of a CDO to be rated, it must at
  the minimum survive the gross default rate at
  that rating level

4
CDO Evaluator

• Analyzes Portfolio Credit Quality
• Creates Probability Distribution of Portfolio
  Default Rates
• Produces the rating Scenario Default Rates needed
  for rating a tranche
• Key Attributes of CDO Evaluator
• Directly incorporates correlation
• No Industry Notching
• Utilizes Monte-Carlo Methodology
• Used for pools of Corporate, ABS, and Sovereign
  debt

5
Portfolio Inputs

• List of Assets in the CDO, specifying
• Obligor Identification number
• Industry or ABS Type
• Principal amount
• Issuer Credit rating
• Maturity
• Sovereign Country and Rating, if needed

6
Sample Asset Input
7
Monte Carlo Methodology

• Precise
• When do enough simulation trials
• Robust
• Need only state the problem
• Can model very complex problems
• Fast
• Todays PCs can do simulations fast

8
Example of Simulation Process
9
Results of 5 Trail Simulation
10
ABS and Corporate Defaults

• Continue modeling Corporate Debt as in the past
  with default rates based on rating and maturity
• Historical ABS Defaults
• Are much less than corporate defaults
• Are not sensitive to Maturity
• Model ABS Securities as
• With its own default rate and independent of
  Maturity

11
Implied Asset Default Rates

• AAA AA A BBB
  BB B
• Corporate
• Year 4 0.19 0.57 0.81 1.81
  9.49 21.45
• Year 7 0.52 1.20 1.81 3.94 14.20
  26.15
• Year 10 0.99 1.99 3.04 6.08
  17.47 28.45
• ABS 0.25 0.50 1.00 2.00
  8.00 16.00

12
Example of Simple Portfolio

• 50 Diverse 10 year BB Corporate Credits
• Probability of Default for each is 17.47 based
  of SPs Corporate default study.

13
Portfolio Default Rate Distribution for Simple
Diverse Portfolio
7 probability of 24 default rate
2.1 probability of Default Rate exceeding 28
14
Next Step Use the Default Distribution To
Determine Scenario Default Rates

• For example, determine the A Scenario Default
  Rate (SDR) such that
• The probability of exceeding it is approximately
  equal to the probability of a A corporate bond
  defaulting.
• Based on default table, 10 yr A credit has
  3.04 probability of default.

15
A Scenario Default Rate

A Corporate Default Probability 3 or
less 28.56 Default rate
16
Scenario Default Rate

• Based on the simulation, under the A
  probability of default is 3.04, the asset pool
  would experience no more than 28.56 default
  rate.
• To rate the CDO tranche A it must them be able
  to withstand 28.56 Gross defaults of the asset
  pool.
• If so, such a tranche should be rated A

17
CDO Tranche Default Probabilities
  Sized for the corporate default tables
AAA AA A BBB BB
B Corporate Year 4 0.19 0.57
0.81 1.81 9.49 21.45 Year 7
0.52 1.20 1.81 3.94 14.20 26.15
Year 10 0.99 1.99 3.04 6.08 17.47
28.45
18
Scenario Default Ratesfor Simple Portfolio
AAA 36.00 AA 33.30 A 28.56 BBB 24.18
BBB
A
AA
AAA
19
ASSUMPTIONS

• Rating of Asset implies its probability of
  default.
• Underlying each Asset is a pool of collateral or
  an obligor whose behavior is correlated.

20
Sector Correlation Assumption

• ABS
• 0.3 within ABS sectors (e.g. CMBS vs. CMBS)
• 0.1 between ABS sectors (e.g. CMBS vs. credit
  cards)
• Corporates
• 0.3 within Industry sectors (e.g. auto vs. auto)
• 0.0 between Industry sectors (e.g. auto vs.
  utilities)

21
AA
AA
22
Step 2 - Synthetic Recoveries

• Absent recoveries and excess spread, each tranche
  would need the level of support equal to the
  gross default rate expected at that rating level.
• Recoveries on Cash Settled synthetics are driven
  by- Mid Point of managed recovery range- Time
  to settlement- of bids and bid taken-
  Cheapest to deliver convertibility option-
  Specified currency- Credit event if old
  restructuring is used

23
Synthetic CDO
Eligible Investments

Assets
cash
Class A Class B Class C Equity
Sponsor/ Protection Buyer
SPV
fee
notes
contingent payment
cash
Reference portfolio
24
Example of Synthetic Recovery

• Mid Point of managed recovery range 37.00
• Cheapest to deliver 2.5 less -
  1.85
• Convertibility option 2.5 less - 0.93
• Specified currency 2.5 less - 0.93
• Time to settlement 50 less - 18.5
• Effective recovery rate 14.8

25
Synthetic Capital Structure

• Evaluator Scenario Default Rates (SDRs)
• AAA Default Rate 40
• BBB Default Rate 20
• Recovery rate
• 25, thus loss is 75 of defaults Losses on
  each rating
• AAA 40 defaults 75 loss 30 BBB
  20 defaults 75 loss 15
• Capital Structure (Tranche size) is
• AAA 70 BBB
  15 Equity (NR) 15

26
Cash Flow CDO Structure

Market / Sponsor
SPV
Loan/Bond Portfolio
Class A Class B Class C Equity
notes
cash
cash
27
Cash Flow Recoveries Based On Assets

• Recoveries on physical debt instruments held by
  the SPV are driven by
• - Insolvency regime- Seniority and security-
  Historical recovery date- Manner of realizing
  recovery- Abilities of Manager

28
Cash Flow CDO Recovery Levels

• Not A Function Of Asset Ratings Or Notes
• Driven By The Seniority Security Of The
  Obligations
• Influenced By The Actions Of The Collateral
  Manager

US Recovery Ranges
Bonds () Loans ()
Senior Secured 40 - 55 50 - 60 Senior
Unsecured 25 - 44 25 - 50 Subordinated 15 -
28 15 - 28
29
Cash Flow Deal Capital Structure
Proposed Capital Structure Is Analyzed by
Modeling

• Default Rate
• Default Timing
• Default Patterns
• Recovery Timing
• Recovery Levels
• Interest Rate Curves
• Hedge Structures
• Asset Characteristics
• O/C I/C testes

Class size at each rating level
Transaction Cash Flow Model
30
Multiple Scenarios In Cash Flows
Fence in as many variables as possible

• 1 Default Rate per Rating
• x 5 Default Timings
• x 5 Default Patterns
• x 5 Interest Rate Curves
• 125 Cash Flow Runs Per Rated Class

31
Sample Results Form Cash Flows

• 5 Loans in Asset Pool - 42.4 Default rate
  needed for AAA
• Maximum Default Rate
• Default Curve 1 Default Curve 2
  Default Curve 3 Default
  Curve 4 Yr 2 Yr 6 Yr 2 Yr 6 Yr 2 Yr 6 Yr
  2 Yr 6
• LIBOR up 62.0 56.5 64.5 58.0 64.0 57.0 61.0
  55.5
• LIBOR Dn 58.5 57.5 59.5 55.5 60.0 57.5 59.0
  58.5
• Swap Rate 53.5 50.5 53.5 49.0 54.5 49.5 54.5
  51.0
• Cap Rate 53.5 50.5 53.5 49.0 54.5 49.5 54.5
  51.0
• Constant 54.0 51.0 54.0 49.5 55.0 50.0 54.5
  51.5
• Up-1 62.5 61.5 65.5 59.0 63.5 62.5 61.0 62
  .0
• Up-2 60.5 49.0 61.5 47.0 61.5 49.0 58.5 50
  .5
• Down-3 56.0 55.0 56.5 53.5 57.0 54.0 56.5 5
  5.5

32
Last, but Not Least

• Legal Analyses - Transfer of Assets
• - SPV Separation
• - Governing Law
• - Transaction Documents
• - Rights of Noteholders

33
Additional Information

• Global Cash Flow and Synthetic CDO Criteria
  March 2002
• Criteria for rating Synthetic CDO Transactions
  Oct. 2003
• Both publications available on
• SPs RatingsDirect,
• www. standardandpoors.com

34
Request Information or CDO Evaluator at

• Local SP Office
• or
• CDOevaluator_at_SandP.com

35
(No Transcript)