SlideShare a Scribd company logo

AIG Conference Call Credit Presentation - Supplemental Materials - February 29, 2008

F
finance2

This document provides an overview of AIG Financial Products' "super senior" credit default swap business. It discusses the business rationale, portfolio composition, underwriting standards, risk assessment, stress testing, accounting and valuation for these types of transactions. Key points include that AIGFP takes a highly conservative approach to underwriting and risk modeling for these exposures, and that even in severe stress scenarios modeled losses are significantly less than current market valuations would imply.

BusinessEconomy & Finance
1 of 158
Download to read offline
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
Conference Call Credit Presentation Supplemental Materials Financial Results for the Year Ended December 31, 2007 February 29, 2008 (Revised as to slide 30 and Capital Markets - Data Appendices)
1 It should be noted that this presentation and the remarks made by AIG representatives may contain projections concerning financial information and statements concerning future economic performance and events, plans and objectives relating to management, operations, products and services, and assumptions underlying these projections and statements. It is possible that AIG's actual results and financial condition may differ, possibly materially, from the anticipated results and financial condition indicated in these projections and statements. Factors that could cause AIG's actual results to differ, possibly materially, from those in the specific projections and statements are discussed in Item 1A. Risk Factors of AIG's Annual Report on Form 10-K for the year ended December 31, 2007. AIG is not under any obligation (and expressly disclaims any such obligations) to update or alter its projections and other statements whether as a result of new information, future events or otherwise. This presentation may also contain certain non-GAAP financial measures. The reconciliation of such measures to the comparable GAAP figures are included in the Fourth Quarter 2007 Financial Supplement available in the Investor Information Section of AIG's corporate website, www.aigcorporate.com.
Capital Markets
3 Outline • Business Rationale, Portfolio Composition & Underwriting Standards • Fundamental Risk Assessment & Stress Testing • Accounting, Valuation Fundamentals & Economic Capital • Conclusions & Next Steps
4 Outline • Business Rationale, Portfolio Composition & Underwriting Standards • Fundamental Risk Assessment & Stress Testing • Accounting, Valuation Fundamentals & Economic Capital • Conclusions & Next Steps
5 • No uniform definition for “Super Senior” risk across the market. • AIGFP defines “Super Senior” risk as a risk where there is no expected loss at contract inception, even under its conservative stress assumptions. • Due diligence and AIGFP proprietary modeling incorporates significantly more conservative assumptions, including for recovery rates, than those used by the rating agencies. • While rating agency models and attachment points are useful verification tools, AIGFP always builds and models each “Super Senior” transaction with its own more conservative assumptions. • The attachment point for the “Super Senior” portion of each portfolio is modeled as a minimum threshold above which there is no expected loss to AIGFP. The final attachment point is negotiated to exceed the modeled attachment point, giving AIGFP an additional cushion of subordination to its risk position. AIG Financial Products “Super Senior” Credit Default Swap (CDS) Business
6 Objectives for Counterparties in “Super Senior” CDS Transactions • Regulatory capital relief • Facilitate securitization • Arbitrage
7 Summary Statistics “Super Senior” Credit Derivatives* Transaction Type Corporate – Regulatory Capital Motivated European Residential Mortgage – Regulatory Capital Motivated Corporate – Arbitrage Motivated Multi-Sector CDOs Transactions w/Mixed Collateral including Subprime Transactions w/No Subprime Total Multi- Sector CDOs Gross Notional ($ Billion) 306.0 182.8 87.3 82.8 27.3 110.1 AIGFP Net Notional Exposure ($ Billion) 229.6 149.1 70.4 61.4 16.8 78.2 Number of Transactions 58 35 36 103 13 116 Weighted Average Subordination (%) ¹ 22.0% 13.8% 18.3% 23.3% 18.0% 21.9% Weighted Average Number of Obligors / Transaction 1,571 74,819 122 194 185 192 Expected Maturity (Years) 1.2 ² 2.3 ² 4.0 5.0 ³ 5.4 ³ 5.1 ³ 1. Weighted by Gross Transaction Notional 2. Maturity shown reflects first non-regulatory call date, although majority of transactions have Regulatory Capital Calls from Jan 08 3. Reflects the Weighted Average Life*All data is as of December 31, 2007.
8 Credit Underwriting – Regulatory Capital Corporate Transactions Every Transaction Passes Through the Same Process of Careful Selection and Due Diligence • For each regulatory capital-motivated transaction, we review the originating bank’s underwriting standards, including for example: – The bank’s internal rating procedures, its construction and criteria, along with its application to both loan and obligor ratings; – The extent to which account officers are empowered to make lending decisions and/or overrule any scoring system; – All loss mitigation and foreclosure strategies; – The bank’s internal rating system to ensure it is seasoned enough to enable it to have built transition matrices to help validate the ratings; and – The mapping of the bank’s internal rating scales to those of the rating agencies and confirmation of the mapping through discussion with the bank and the rating agencies. • Investigation of any industry or geographic concentrations in the loan pool and review of rationale and any potentially mitigating factors. We strive to create diverse and granular pools of credits. • Where possible, we undertake an analysis of each name in the portfolio to assign each an internal rating (‘AIGFP rating’). The AIGFP rating (which in all cases is equal to or lower than those publicly assigned) is used for modeling purposes. • Price negotiation is bespoke and reflects the counterparty’s unique position and capital requirements. With the implementation of Basel II, such pricing typically reflects an agreed minimum fee due to the very short expected life of the transaction.
9 Credit Underwriting – Regulatory Capital European Residential Mortgage Transactions • Understanding of the bank’s lending philosophy and the extent (if any) to which this has changed in recent times • Underwriting standards and credit approval procedures, including: – How many and which individual officers are empowered to make lending decisions – Loan size that can be approved by different individuals/groups – The extent to which the approval system has been automated and, if so, the procedures for permitting any overrides – What criteria do the credit officers or automated system use to make the lending decision • Loss and recovery experience of the bank • Historical experience - annual losses over the past 10 years or as far back as the bank’s records allow • Monitoring of delinquencies, work-out and recovery procedures • Groups and procedures in place to monitor and manage delinquencies • Bank policies on work-out and foreclosure • Price negotiation is bespoke, similar to regulatory capital corporate transactions. Thorough Review of all Underwriting, Processes and Performance
10 Credit Underwriting – Arbitrage Motivated Corporate Transactions Every Transaction Passes Through the Same Process of Careful Selection and Due Diligence • Similar to the Regulatory Capital transactions, an internal AIGFP rating is assigned to each name in the portfolio (which in all cases is equal to or lower than those publicly assigned) and used for modeling purposes. • Removal of any stressed credits or those which, in our view, have particularly unfavorable outlooks. • Review of current market spreads for each name, where available, to ensure rating used reflects all currently available information. • Selection of credits so as to ensure a diverse and granular pool both in terms of industry and geography. • Pricing discussion based around both the current spread environment and the counterparty’s specific hedging requirements at that particular time.
11 Past Performance Corporate Loans and European Residential Mortgages AIGFP Net Notional Exposure ($ Billions) Total Losses in Reference Pool to Date Weighted Average Attachment Point Multiple of Losses Required Before AIGFP Has any Loss1 Corporate Loans – Regulatory Capital $229.6 0.01% 22.0% 2,234x European Residential Mortgages – Regulatory Capital $149.1 0.03% 13.8% 487x Corporate Loans – Arbitrage $70.4 0.22% 18.3% 81x • The losses realized to date in the underlying reference pools have been extremely small relative to our attachment points. This is a reflection of the positive selection of the portfolios, the parties’ motivations and the highly conservative modeling. AIGFP has not incurred any realized losses from the underlying collateral in these pools. December 31, 2007 Extremely Low Loss Rates in Reference Pools with Attachment Points Always Significantly Higher 1.1. Represents multiple of total losses in reference pool to date reRepresents multiple of total losses in reference pool to date required to exhaust the entire subordination structure before AIGFquired to exhaust the entire subordination structure before AIGFP incurs a loss.P incurs a loss.
12 Any realized credit losses are allocated sequentially: Equity, BB, BBB, A, AAA, then “Super Senior” Underlying portfolio typically comprises 125-200 obligations from various sectors. Those obligations typically have their own subordination embedded “Super Senior” Risk Layer - AIGFP Net Notional Exposure Typical Tranche Structure of a Multi-Sector CDO Including “Super Senior” Layer Equity BB BBB A AAA AAA A AA BBB AAA A AA BBB AAA A AA BBB AAA A AA BBB AAA A AA BBB AAA A AA BBB AAA AA A BBB AAA AA A BBB AAA AA A BBB AAA AA A BBB AAA AA A BBB Portfolio tranched into different risk layers AIGFP Attachment Point Gross Transaction Notional Residential and commercial mortgages, auto loans, etc., are securitized Specific individually rated tranches from those securitizations are purchased by the CDO The CDO is tranched into different layers of risk with the “Super Senior” layer being the most risk remote Protection buyer makes periodic payments to protection seller who in turn makes payments if losses, which are allocated sequentially, exceed the relevant subordination
13 • Review and analysis of the CDO Manager (on all transactions), including: – Review of personnel and their experience and suitability for managing the assets and the structure – Track record and past performance of the manager in all asset classes – Risk retention / incentive policy in place for key employees • Review and analysis of the entire collateral portfolio, including: – The eligibility criteria for all securities – The proposed single security / obligor concentration limits – Geographic portfolio diversification – Sector / industry portfolio diversification – Maturity / expected amortization profile of the assets and the portfolio – Review of agency ratings of securities and portfolio weighted average rating factor – Currency and interest rate exposures and hedging requirements Credit Underwriting – Multi-Sector CDO Transactions Highly Selective Review of Manager, Collateral and Proposed Structure
14 • Analysis of the key transaction terms, including: – The term of any proposed re-investment period – Management trading discretions, if any – Portfolio quality triggers in place – Over-Collateralization (O/C) and Interest Coverage (I/C) tests – Early amortization events and required procedures Credit Underwriting – Multi-Sector CDO Transactions Highly Selective Review of Manager, Collateral and Proposed Structure
15 Multi-Sector CDOs What Differentiates our Transactions? • Strong adherence to proven conservative underwriting approach. • Highly diversified and granular pools of positively selected reference obligations. • Attachment points are calculated to be extremely remote, frequently with significant AAA-rated tranches below our position. • Calculated attachment points are only a minimum and are non-negotiable. • Extensive due diligence carried out before any transaction is agreed. • Conservative portfolio and obligor limits. • Conservative modeling – Conservative assumptions used for portfolio construction – Significant hair-cuts and stresses applied to inputs Careful Portfolio Selection Combined with Remote Attachment Points
16 Outline • Business Rationale, Portfolio Composition & Underwriting Standards • Fundamental Risk Assessment & Stress Testing • Accounting, Valuation Fundamentals & Economic Capital • Conclusions & Next Steps
17 Fundamental Risk Assessment: AIGFP Credit Review Process • AIGFP’s highly experienced credit team as part of its role conducts continual surveillance around the Super Senior portfolio. • Each quarter AIGFP re-runs its stress models against the entire Super Senior portfolio to account for updated subordination information, ratings migration, delinquencies, defaults and losses. They update, evaluate and stress results relative to the current subordinated layers to assess potential credit quality migration. • The AIGFP global credit team meets quarterly to review the portfolio in depth and the results of the stress model. • They review any deals that show early signs of stress and evaluate the factors leading to any portfolio deterioration to determine whether exposure hedging should be recommended or other actions should be instituted, such as meetings with collateral managers or bank lenders. • The credit review also looks at rating agency changes, early deal terminations and credit trends.
18 Fundamental Risk Assessment AIG ERM • Every quarter AIG ERM reviews AIGFP’s “Super Senior” credit derivative exposures. • The review considers delinquency, defaults and realized loss trends for each transaction relative to updated subordination levels. The assessment includes a review of rating agency actions. It also considers adverse economic and sector trends, where applicable. • ERM identifies all transactions that show any unexpected deterioration or heightened risk and adds them to the internal AIG Watch List. • ERM has initiated a regular process to run stress tests of the multi-sector CDO portfolio to determine if any transactions could pose a risk of realizing a loss if economic conditions deteriorate beyond expectations. • ERM also assesses whether any transactions could represent probable loss, thus potentially requiring the establishment of credit reserves (none to date).
19 Rating Agency Actions • All three major rating agencies continue to rate 89.6% of AIGFP’s $61.4 billion super senior credit derivative multi-sector CDO portfolio with sub-prime RMBS collateral at AAA levels. This is despite massive numbers of CDO downgrades during 2007 and early 2008. • Through February 26, 2008 and based on 12/31/07 multi-sector CDO exposure data, approximately $6.4 billion (10.5%) of the portfolio had been downgraded, mostly by only one agency, as follows (using the lowest rating of any of the three agencies): – $2.3 billion to AA grades; – $2.0 billion to single A grades; and – $2.1 billion to BBB grades • At least one rating agency continues to rate all $6.4 billion downgraded CDO SS tranches as AAA. • Another $8.7 billion of AAA rated super senior CDOs (twelve transactions) currently are on credit watch, mostly by S&P. • No exposure is rated below BBB-.
20 Stress Testing - Illustration of Potential Losses on AIGFP’s “Super Senior" Credit Derivative Portfolio on Multi-Sector CDOs 0.90 11.25 0.00 2.00 4.00 6.00 8.00 10.00 12.00 Modeled Severe Stress Scenario Realizable Loss Unrealized Market Valuation Loss Carried on GAAP Balance Sheet The December 31, 2007 unrealized market valuation loss of $11.25 billion significantly exceeds even a severe modeled realizable portfolio loss. Value of Pre Tax Loss Estimates* $ BN Description of ERM Severe Stress Scenario* Collateral Securities Severe Stress Scenario Q1-Q4 ’07 Subprime RMBS 100% of AA+ or lower Q3-Q4 ’06 Subprime RMBS 100% of AA+ or lower Q1-Q2 ’06 Subprime RMBS 50% of AA+, AA, AA-; 100% of A+ or lower Q3-Q4 ’05 Subprime RMBS 50% of BBB+ or lower Q1-Q2 ’05 Subprime RMBS 100% of BB+ or lower Inner CDOs of ABS 100% of A+ or lower CY’06 & CY’07 Alt-A 100% of A+ or lower *As of December 31, 2007. These stresses are “static” stresses, assumed to result in immediate portfolio loss and do not take any benefit for cash flow diversion and other mitigants.
21 Market Valuation vs. Stress Testing Illustrations • Portfolio Level – Slide 21 shows that market valuations indicate losses that far exceed severe stress scenario realizable losses at total portfolio level. • Transaction Level – Slides 22 to 24 show loss comparisons at transaction level: In all CasesIn all Cases Market valuations indicate losses that exceedMarket valuations indicate losses that exceed severe stress scenario realizable losses forsevere stress scenario realizable losses for each of the transactionseach of the transactions Market valuations indicate losses while severeMarket valuations indicate losses while severe stress scenario realizable losses, generally dostress scenario realizable losses, generally do not even breach the subordination layersnot even breach the subordination layers -- ILLUSTRATIONILLUSTRATION -- -- ILLUSTRATIONILLUSTRATION -- 0.0 1.0 2.0 $Billions In ManyIn Many CasesCases0.0 1.0 2.0 $Billions In Most CasesIn Most Cases AIGFP Net Notional Exposure Subordination Layer Severe Stress Scenario Realizable – Gross Notional Loss Unrealized Market Valuation Loss
22 Stress Testing - Illustration of Potential Losses on AIGFP’s “Super Senior” Credit Derivative Portfolio on Multi-Sector CDOs $ Billions Total AIGFP Net Notional Exposure: $78.2 BN Unrealized Market Valuation Loss: $11.25 BN Severe Stress Realizable Loss: $0.90 BN Attachment Point Subordination Layer “Super Senior” Total Multi-Sector CDO Portfolios Unrealized Market Valuation Loss Severe Stress Scenario Realizable Loss
23 Multi-Sector CDO Portfolio Transactions 0.0 1.0 2.0 3.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 $Billions AIGFP Net Notional Exposure Subordination Layer Severe Stress Scenario Realizable - Gross Notional Loss Unrealized Market Valuation Loss Transaction by Transaction Illustration
24 Multi-Sector CDO Portfolio Transactions 0.0 1.0 2.0 3.0 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 $Billions AIGFP Net Notional Exposure Subordination Layer Severe Stress Scenario Realizable - Gross Notional Loss Unrealized Market Valuation Loss Transaction by Transaction Illustration
25 Multi-Sector CDO Portfolio Transactions 0.0 1.0 2.0 3.0 4.0 5.0 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 $Billions AIGFP Net Notional Exposure Subordination Layer Severe Stress Scenario Realizable - Gross Notional Loss Unrealized Market Valuation Loss Transaction by Transaction Illustration
26 Current Market Implied Probabilities of Default for AIGFP’s “Super Senior” Credit Derivative Portfolio on Multi-Sector CDOs 53% 26% 2%0% 10% 20% 30% 40% 50% 60% Estimate Derived From Historical Default Probabilities Current Exit Value Implied By Market Prices Severe Stress Scenario Implied Probabilities of Default of Gross Exposures Over Lifetime of Portfolio* Dec 2006 Dec 2007 Dec 2007 Market Implied Probabilities of Default Reflect: • Basic Credit Risk Fundamentals; plus • Extreme Liquidity Premium; • Market-Driven Risk Aversion; etc. Current market implied probabilities of default, being orders of magnitude greater than historical probabilities of default, suggest the existence of significant factors included in market prices in addition to credit risk, the principal risk to which AIGFP is exposed. * Implies a weighted average recovery rate of 40% across all por* Implies a weighted average recovery rate of 40% across all portfolios.tfolios.
27 Conclusions From Fundamental Risk Assessment & Stress Testing • All three major rating agencies continue to rate 89.6% of AIGFP’s $61.4 billion super senior CDS multi-sector CDO portfolio with sub-prime RMBS collateral at AAA levels. This is despite massive numbers of CDO downgrades during 2007 and early 2008. No exposure is rated below BBB-. • The December 31, 2007 unrealized market valuation loss of $11.25 billion significantly exceeds even a severe modeled realizable portfolio loss. • Market valuations indicate losses that exceed severe stress scenario losses for each of the transactions. For most of the transactions, market valuations indicate losses, while severe stress scenario realizable losses generally do not even breach the subordination layers. • Current market implied probabilities of default, being orders of magnitude greater than historical probabilities of default, suggest the existence of significant factors included in the market prices in addition to credit risk, the principal risk to which AIGFP is exposed. • AIGFP wrote credit derivative protection as a principal based upon sound underwriting procedures and has the ability and intent to hold its positions until contract maturity or call by the counterparty.
28 Outline • Business Rationale, Portfolio Composition & Underwriting Standards • Fundamental Risk Assessment & Stress Testing • Accounting, Valuation Fundamentals & Economic Capital • Conclusions & Next Steps
29 Accounting for “Super Senior” Credit Derivative Swaps • AIGFP accounts for its “Super Senior” Credit Derivative portfolio in accordance with FAS 133 and EITF 02-3: – At inception the credit derivative is recorded at its transaction price as that is the best indicator of fair value. – Subsequent changes in fair value are recognized in earnings. • Through June 30, 2007 there was minimal change in fair value since the inception of the derivatives: – The “Super Senior” credit derivative transactions are significantly out-of-the-money put options that are insensitive to normal changes in market credit spreads. – A significant change in credit spreads is required to cause a material change in fair value. Credit spread changes did not result in a significant change to fair value losses until the third and particularly the fourth quarters of 2007.
30 AIGFP “Super Senior” CDS Portfolio Total Notional Amounts and Cumulative MTM Loss Type Notional Amount ($ Billions) Cumulative MTM Loss ($ Billions) Corporate Arbitrage $ 70.4 $ 0.2 Regulatory Capital* $ 378.7 $ 0.0 Multi-Sector CDO, of which: High Grade Mezzanine $ 78.2** $ 59.3 $ 18.9 $ 11.3*** $ 7.1 $ 4.2 Total: $ 527.3 $ 11.5*** (December 31, 2007) * Represents Corporate & European Residential Mortgage Regulatory Capital transactions. ** The average amount of defaulted collateral in the multi-sector CDOs is 51 basis points. There are four outliers to this average where defaulted collateral exceeds 10%, the maximum of which is 13.2%. However, in all these cases the subordination is in excess of 43% with the maximum being 64.4%. *** Includes benefit of $310 million attributable to cash flow diversion features.
31 Evolution of AIGFP’s Valuation Methodologies for “Super Senior” Corporate Arbitrage Transactions • At December 31, 2007, AIGFP valued its Corporate Arbitrage “Super Senior” credit derivative transactions using relevant market indices or third party prices. • AIGFP reported a mark-to-market loss in the amount of $226 million under this approach in the fourth quarter of 2007. • At September 30, 2007 AIGFP employed the BET model to value this portfolio, resulting in no noticeable change in fair value.
32 Regulatory Capital (Corporate - Regulatory & European Residential Mortgage) “Super Senior” Transactions Valuation • Transactions entered into are of a highly customized, non-market standard nature, to facilitate regulatory capital relief, rather than for credit risk transfer. • Transactions are expected to terminate in conjunction with the implementation of Basel II (within 12 to 18 months). • AIG conducted a comprehensive analysis of information available at year end, including counterparty motivation, portfolio performance, market place indicators and transaction-specific considerations. • The most compelling market observable data is the termination of $54 billion of transactions in early 2008. AIG was not required to make any payments and was paid a termination fee in some terminations. • Hence AIG believes that these regulatory trades are appropriately valued at zero fair value as of December 31, 2007.
33 Evolution of AIGFP’s Valuation Methodologies/Inputs for “Super Senior” Credit Derivatives Written on CDOs Date September 30, 2007 October 31, 2007 November 30, 2007 (Method A) November 30, 2007 (Method B) December 31, 2007 Methodology • Modified BET • No attribution of Cash Flow Diversion (CFD) • Modified BET • No attribution of Cash Flow Diversion (CFD) • Modified BET • Attribution of Cash Flow Diversion (CFD) using Monte Carlo simulation • Modified BET • Attribution of Cash Flow Diversion (CFD) using Monte Carlo simulation • Negative Basis Adjustment • Modified BET • Attribution of Cash Flow Diversion (CFD) using Monte Carlo simulation • No Negative Basis Adjustment • Overlay of Super Senior Tranche Price Quotes Inputs • Third party credit spreads on generic ABS • Moody’s recovery rates • Third party credit spreads on generic ABS • Third party spreads on RMBS collateral adjusted for relative change in ABX.HE • Moody’s recovery rates • Third party credit spreads on generic ABS • Third party spreads on RMBS collateral adjusted for relative change in ABX.HE • Moody’s recovery rates • Third party prices collected by CDO managers during November for October month- end • Moody’s recovery rates • Third party prices collected by CDO managers during January for December month- end • Moody’s recovery rates
34 Process Followed for December 31, 2007 GAAP Valuation of “Super Senior” CDS Written on CDOs Acquisition &Acquisition & Review of ThirdReview of Third Party Prices ofParty Prices of CollateralCollateral SecuritiesSecurities BenchmarkingBenchmarking to Independentto Independent SourcesSources ModeledModeled SuperSuper SeniorSenior marketmarket value lossvalue loss Overlay ofOverlay of Super SeniorSuper Senior Tranche PriceTranche Price QuotesQuotes Acquisition &Acquisition & Review of ThirdReview of Third Party Prices ofParty Prices of UnderlyingUnderlying Securities ObtainedSecurities Obtained Through CDOThrough CDO Managers:Managers: •• Obtained dealerObtained dealer prices on 70% ofprices on 70% of securities of allsecurities of all portfoliosportfolios combinedcombined •• Derived finalDerived final price byprice by averaging inaveraging in case of multiplecase of multiple quotesquotes •• Reviewed pricesReviewed prices for consistencyfor consistency across ratingsacross ratings and timeand time Benchmarking ofBenchmarking of Third Party Prices toThird Party Prices to Independent PriceIndependent Price Sources:Sources: •• To IDC pricesTo IDC prices (9,180(9,180 securities);securities); •• BloombergBloomberg –– although feweralthough fewer matches (1,124matches (1,124 securities);securities); •• Monthly trendsMonthly trends in ABX.HE indexin ABX.HE index (Series 6(Series 6--1, 61, 6--2,2, 77--1)1) Key InputsKey Inputs to Modifiedto Modified BET ModelBET Model Acquisition andAcquisition and review of other keyreview of other key inputs to theinputs to the Modified BETModified BET model:model: •• WAL ofWAL of securitiessecurities -- Bloomberg;Bloomberg; •• Verification ofVerification of WAL usingWAL using prepaymentprepayment model;model; •• Use of matrixUse of matrix pricing;pricing; •• Diversity score;Diversity score; •• LIBOR curve forLIBOR curve for discounting cashdiscounting cash flows;flows; •• Recovery ratesRecovery rates based onbased on MoodyMoody’’s multis multi-- sector CDOsector CDO recovery datarecovery data Valuation, reviewValuation, review and stress testing ofand stress testing of Modified BETModified BET results of the Superresults of the Super Senior marketSenior market valuation loss:valuation loss: •• Convert price toConvert price to spread;spread; •• Use key inputsUse key inputs to run BET;to run BET; •• Apply OC testsApply OC tests and implementand implement CFD algorithms;CFD algorithms; •• Stress testingStress testing inputs andinputs and validation usingvalidation using separateseparate GaussianGaussian Copula model;Copula model; •• Validation byValidation by applying modelapplying model to observedto observed CDX SS pricingCDX SS pricing Overlaying theOverlaying the Super SeniorSuper Senior Tranche QuotesTranche Quotes Obtained From 12Obtained From 12 Major Dealers to theMajor Dealers to the modified BET modelmodified BET model resultsresults
35 “Super Senior” Multi-Sector CDO Valuation Model • The Binomial Expansion Technique (BET) methodology was originally developed by Moody’s for rating portfolio credit products and is transparent, widely used by market participants and in the public domain. • We modified it to imply default probabilities from market prices for the underlying securities, not from rating agency assumptions. • The model replaces a large collateral pool of correlated assets with a smaller pool of idealized homogeneous, independent assets • The size of the idealized pool, i.e., the number of assets, is given by the Diversity Score. • The BET model placed in a Monte Carlo simulation framework enables AIGFP to: – Derive a loss distribution through time for the portfolio; – Value the important structural features of each transaction.
36 “Super Senior” Multi-Sector CDO Valuation Model Model parameters are derived from independent market sources. • Third party prices for the underlying securities that comprise the collateral of each CDO are obtained from CDO managers. • AIGFP was able to obtain prices for 69% of unique collateral securities comprising the underlying collateral. • From these prices AIGFP derives credit spreads and market-implied default probabilities. • Diversity Scores are generally provided by CDO trustees as the determinant of correlation. • Weighted Average Life for underlying securities are obtained from third-party data providers. • Assumed Recovery Rates for each underlying security is obtained from Moody’s historical experience.
37 “Super Senior” Multi-Sector CDO Valuation Model • The BET model has been reconciled by a separate AIGFP team using a Gaussian Copula model. • Valuation results were consistent when the model’s inputs were calibrated consistently. • Both models confirmed via parameter stress testing that in the current market environment valuations had a relatively low sensitivity to: – the correlation input (diversity score), and – the loss distribution function implied by each model.
38 Negative Basis (1 of 2) • Negative basis refers to the fact that cash and synthetic instruments can often trade at different levels, with cash instruments at a wider spread than CDS. • Cash instruments trade at a spread above the risk-free rate. The spread incorporates the risk of default (credit risk), the market’s price for liquidity risk and, in distressed markets, risk aversion costs. • One component of negative basis is the funding cost. A CDS, i.e., a synthetic position, does not need to be funded initially and hence saves the investor the cost of financing the position and potentially any balance sheet usage. – A typical example: • A cash bond for corporate XYZ trades at a spread of Libor + 250 bps • The credit protection on the same credit to the matching maturity is offered at a spread of Libor +150 bps • In the trade the investor buys both the bond and the credit protection, removing any credit risk embedded in the bond and locking in a positive spread of 100 bps • This funding cost, which is reflected in the market’s price for liquidity risk of cash instruments, is clearly greater in times of crisis or market illiquidity, as the benefit of holding cash increases and the negative basis also tends to rise.
39 Negative Basis (2 of 2) • There is a well established market in corporate credit with investors buying bonds and simultaneously buying matching credit protection on the same instrument in order to “lock in” the positive spread differential. This is termed the negative basis. – There is an active investment grade corporate market to transact, collect market quotes and document its existence. • The negative basis in the structured credit market follows the same principles and exists for the same reasons as in the corporate credit market. – As a result of the ABS market currently being more illiquid, collecting sufficient market quotes and documenting the existence of the negative basis in the structured credit market is problematic. • Our transactions are evidence of an observable negative basis market at inception on CDOs of ABS. • There are other components which also drive the basis, some of which make it less negative or may even make it positive at times.
40 Why Did We Not Incorporate Negative Basis into Our Valuations as of December 31, 2007? • AIGFP did not adjust its fair valuations for the negative basis, even though it believes such a difference exists. • AIG must collect sufficient observable evidence that supports existence and realization of negative basis to permit AIG to factor negative basis into its valuation methodology. • AIG is not able to obtain sufficient objective evidence at this time as a result of the market dislocation and lack of trading.
41 Similarities Between Excess Casualty Insurance & AIGFP’s “Super Senior” Credit Derivative Portfolio on Multi-Sector CDOs • AIG acts as principal • AIG retains underwriting control • Relatively high attachment points • Gives AIG access to specialized (non-commodity) markets • AIG can “pick and choose” risks – broker has no binding authority • Liabilities are generally not traded, but held to settlement • Reinsurance used selectively to take advantage of market pricing Excess Casualty Insurance “Super Senior” Credit Derivative Portfolio on Multi-Sector CDO • AIG acts as principal • AIG retains underwriting control • Relatively high attachment points • Gives AIG access to specialized (non- commodity) markets • Largely bespoke transactions with tailored contractual terms • “Super Senior” credit derivatives are generally not traded, but held to settlement • Protection purchased selectively for rated layers on portfolios of reference obligations
42 Valuation Principles Applied By AIG • “The price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date” • “NOT a forced transaction (for example, a forced liquidation or distressed sale)”; but • Represents current “exit price” in the principal market “Fair Value” Under U.S. GAAP “Market Consistent Settlement Value” under AIG’s Economic Capital Model (ECM) • For tradeable assets and liabilities, represents the market value without adjustment for “own credit standing” • For non-tradeable assets and liabilities (e.g., those requiring use of “Level 3” inputs under FAS 157), represents current estimated value of asset or liability plus a risk margin for bearing risk out to settlement • No “exit” assumed at valuation date – avoids imputing characteristics of a liquidation in times of market stress Source: Centre for Audit Quality (CAQ) , White Paper, “Measurements of Fair Value in Illiquid (or less liquid) Markets”; October 3, 2007 (www.aicpa.org)
43 Illustration of Key Differences Between Fair Value & Market Consistent Settlement Value for Non-Tradeable Assets/Liabilities • Most significant difference is in risk and service margin component • Under Market Consistent Settlement Value, an allowance is made for the cost of bearing risk (only) until settlement of the liabilities using a market standard technique (e.g., Cost of Capital) • Under current exit value, the “risk & service margin” may be materially inflated in times of market crisis for the opportunity of a buyer of the liabilities to recover a super normal return for : – Liquidity Risks – Sunk Costs & Opportunity Costs – Return on Franchise Value – Supply/Demand Imbalance Advantages Example: Valuation of Non-Tradeable Liability Consequent to Liability Market Crisis (e.g. Post KRW in P&C Insurance) Fair Value Based on Current Exit Prices (e.g. per FAS 157) Market Consistent Settlement Value Risk & Service Margin Remove Own Credit Risk Adjustment Current Discounted Estimate of Contractual Cash Flows Market Consistent Settlement Value is more appropriate for determining the economic position of AIG, as AIG generally intends to, or is obligated to, hold its illiquid asset and liability positions until maturity. -ILLUSTRATION -
44 Determination of Market Consistent Settlement Value of AIGFP’s “Super Senior” Credit Derivative Portfolio on Multi-Sector CDOs for AIG’s Economic Capital Model (December 31, 2007) 1.71 0.81 0.90 0.00 0.50 1.00 1.50 2.00 2.50 Estimated Realizable Loss Consequent to Severe Stress Cost of Capital Risk Margin Market Consistent Settlement Value Loss Under AIG's ECM AIG’s conservative estimate of Market Consistent Settlement Value Loss for Determining Available Economic Capital: 1) Realizable losses associated with the “severe” stress scenario over-ride best estimate loss (allows for possible change in credit risk fundamentals) 2) A risk margin is added to the component under 1), allowing for the cost of servicing capital requirements* 3) Market Consistent Settlement Value is the sum of 1) and 2) above Conservative Estimate of Market Consistent Settlement Value Loss Under AIG’s Economic Capital Model (ECM) * Determined using the Cost of Capital approach recommended by the CRO Forum for non-tradeable/ non- hedgeable risks (refer to www.croforum.org) $ BN$ BN
45 Adjustments Made to Unrealized GAAP Market Valuation Loss to Determine Available Economic Capital Under AIG’s ECM 1.71 9.54 6.20 11.25 0.00 2.00 4.00 6.00 8.00 10.00 12.00 Unrealized Market Valuation Loss Carried on GAAP Balance Sheet Market Consistent Settlement Value Loss Pre-Tax Adjustment to Unrealized Losses After-Tax Adjustment to GAAP Equity • At June 30, 2007 AIG used GAAP equity as a conservative proxy for Available Economic Capital • For December 31, 2007: – AIG will use Market Consistent Embedded Value* as its estimate of Available Economic Capital for the Life & Retirement Services segment – For the General Insurance segment, a consistent approach will be used – These valuation approaches are consistent with the market consistent settlement value approach AIG has applied to FP’s Super Senior credit derivative portfolio of Multi-Sector CDOs Market Consistent Settlement Value Adjustment to Determine Available Economic Capital (December 31, 2007) $BN * Currently being independently reviewed and certified by Towers Perrin. For the purposes of determining Available Economic Capital, AIG believes it is reasonable to make a positive market consistent settlement value adjustment of $6.2 billion in respect of the AIGFP Unrealized Loss to its GAAP Reported Total Shareholders’ Equity as at December 31, 2007.
46 Outline • Business Rationale, Portfolio Composition & Underwriting Standards • Fundamental Risk Assessment & Stress Testing • Accounting, Valuation Fundamentals & Economic Capital • Conclusions & Next Steps
47 Conclusions • AIGFP “Super Senior” credit derivative business was underwritten to an AIGFP super senior standard which projects zero losses at inception. • There has been deterioration in the credit quality and the market’s assessment of expected losses in the underlying collateral securities and AIGFP’s “Super Senior” credit derivative portfolios. • In accordance with GAAP, AIG recognized a sizable unrealized market valuation loss in 2007, consequent to severe market disruption and credit deterioration, particularly in US sub-prime mortgages, occurring predominantly in the fourth quarter. This market adjustment represented management’s best estimate of the exit value of this portfolio into the current illiquid and distressed market. This volatile market may persist for some time. • Despite the unrealized market valuation loss: – Based upon its most current analyses, AIG believes that any losses AIGFP may realize over time as a result of meeting its obligations under these derivatives will not be material to AIG’s consolidated financial position, although it is possible that such realized losses could be material to AIG’s consolidated results for an individual reporting period. – Except to the extent of credit impairment losses, AIG expects the unrealized market valuation losses to reverse over the remaining life of the portfolio. • AIGFP wrote credit protection as a principal based upon sound underwriting procedures and has the ability and intent to hold its positions until contract maturity or call by the counterparty.
48 Next Steps • Remediate the material weakness in internal control over financial reporting with respect to the controls over the AIGFP “Super Senior” credit derivative portfolio valuation process and oversight thereof. • Continue to monitor closely the portfolio risk of realized loss through fundamental analysis, modeling and stress testing. • Enhance current valuation methodologies and processes – Improve the timeliness and comprehensiveness of data inputs – Consider/develop/implement additional modeling techniques – Continue to observe the market • Pursue opportunities in the market.
Insurance Investments
50 AIG Insurance Investment Portfolios • The investment portfolios of AIG’s insurance companies are managed by AIG Investments (AIGI)* on their behalf. • These portfolios are managed on a spread investment or Asset-Liability Management model, not as a transactional business. As a result, we do not: – “Warehouse” residential mortgage loans or securitizations; or – Retain residual or other securities from residential mortgage backed securities (RMBS) activities. • AIGI’s RMBS and commercial mortgage backed securities (CMBS) are predominantly held as “Available for Sale” securities, not as trading positions. Hence, our underwriting focus is on ultimate collectibility, not short-term market movements. • AIG, as with all investments, purchases RMBS, Asset-Backed Commercial Paper, Structured Investment Vehicles, and RMBS-based Collateralized Debt Obligations (CDOs) based on proprietary internal research. • All information presented in this document is as of December 31, 2007, unless otherwise noted. • All figures are based on amortized cost** unless otherwise indicated. • Ratings used in this presentation are external ratings, or equivalent, based on AIG’s internal risk rating process. * For purposes of this presentation, AIGI is used to denote AIG Insurance Investment Portfolios. ** Amortized cost is the cost of a debt security adjusted for amortized premium or discount less other-than-temporary impairments.
51 AIG Insurance Investment Portfolios Worldwide Insurance and Asset Management Bond Portfolios • AIGI’s bond portfolios* had a fair value of $491.3 billion at December 31, 2007. • The securities are highly rated (approximately 95% are investment grade). • The bond portfolios are also well-diversified geographically. * Fixed Maturities: Bonds available for sale, Bonds held to maturity, Bonds trading securities and Bonds available for sale included in Securities lending collateral $286.7 Billion $204.6 Billion Foreign Bonds by Ratings Domestic Bonds by Ratings AA 44% A 27% AAA 21% Lower 2% BBB 6%BBB 15% Lower 7% AAA 51% A 12% AA 15%
52 AIG Insurance Investment Portfolios Domestic Bonds by Category Other ABS, 2% CDO Debt, 1% Credit, 42% CMBS, 4% RMBS, 27% U.S. Government, 1% Municipal, 23% $286.7 Billion
53 AIG Insurance Investment Portfolios Accounting and Valuation • AIGI accounts for its RMBS, CMBS and CDOs in accordance with FAS 115, FSP FAS 115-1 and EITF 99-20. – These securities are predominantly classified as available for sale securities under FAS 115. – Changes in fair value are reported in other comprehensive income, net of tax, as a component of shareholders’ equity until realized. – Realization through earnings occurs when the position is either sold or is determined to be other-than-temporarily impaired. • AIGI utilizes external pricing vendors as a primary pricing source. – 95% of the portfolio fair values are derived from prices provided by industry standard commercial pricing vendors – such as IDC, Bloomberg and Lehman Brothers. – Vendor pricing methodology and broker prices are internally reviewed for reasonableness, but these securities are not valued solely based on internal models. • The valuation of these securities varies by the type of collateral, the position in the capital structure and the vintage.
54 AIG Insurance Investment Portfolios Other Than Temporary Impairments (OTTI) • AIG’s senior management evaluates its investments for impairment such that a security is considered a candidate for other-than-temporary impairment if it meets any of the following criteria: – Trading at a significant (25 percent or more) discount to par, amortized cost (if lower) or cost for an extended period of time (nine consecutive months or longer); – The occurrence of a discrete credit event resulting in (i) the issuer defaulting on a material outstanding obligation; (ii) the issuer seeking protection from creditors under the bankruptcy laws or any similar laws intended for court supervised reorganization of insolvent enterprises; or (iii) the issuer proposing a voluntary reorganization pursuant to which creditors are asked to exchange their claims for cash or securities having a fair value substantially lower than par value of their claims; or – AIG may not realize a full recovery on its investment, regardless of the occurrence of one of the foregoing events. • An impairment charge may also be taken in light of a rapid and severe market valuation decline because AIG could not reasonably assert that the recovery period would be temporary (severity losses). • AIG Investment’s Chief Investment Officer – Insurance Companies and Chief Credit Officer make credit- related OTTI recommendations using three categories: a) likely to recover; b) possible to recover; and c) unlikely to recover, based on a detailed written description of the circumstances of each security. • In addition, in accordance with EITF 99-20 an analysis of the anticipated cash flows supporting each asset backed security (ABS), representing rights to receive cash flows from asset pools, such as CDOs, RMBS, CMBS, etc., and generally rated below AA-, is prepared and reviewed for impairment. • All credit-related OTTI recommendations, together with supporting documentation, are reviewed on a quarterly basis and approved by AIG’s Chief Credit Officer. The AIG Chief Credit Officer must also determine whether there are any additional securities (not on the list submitted by AIG Investment’s Chief Investment Officer – Insurance Companies) that should be written down.
55 AIG Insurance Investment Portfolios Consolidated Summary of Gains & Losses Financial Effect of Market Disruption For the Year Ended December 31, 2007 Realized and Unrealized Gains / Losses (Pre-tax) ($ Million) Total AIG* Amount Attributable to RMBS Portfolio Net realized capital gains (losses) ($3,592) ($1,647) of which, Sales Activity $1,237 ($30) OTTI ($4,072) ($1,617) Other** ($757) ($0) Unrealized (depreciation) appreciation of investments (included in Other comprehensive income) ($8,046) ($5,070) of which, AAA-rated RMBS (depreciation) ($4,633) ($4,633) AA-rated RMBS (depreciation) ($752) ($752) Lower than AA-rated RMBS (depreciation) ($118) ($118) RMBS appreciation $433 $433 • The other-than-temporary impairments and unrealized losses result primarily from the severe credit and liquidity market turmoil. * Excludes AIGFP’s super senior credit default swap portfolio. **Consists predominantly of foreign exchange related losses.
56 AIG Insurance Investment Portfolios RMBS Portfolios
57 AIG Insurance Investment Portfolios RMBS Overview • Non-agency RMBS are issued in tranche structures, such that the lower tranches absorb losses on the underlying collateral in the pool and thus insulate the higher rated tranches from loss. – The structure and size of each tranche depend on the nature of the collateral, rating agency analysis and models of default scenarios. – As a general rule, AAA and AA rated securities are structured to withstand default losses within the collateral that are multiples of historical norms without any loss of principal or interest. • Holdings of global residential mortgage market products total approximately $89.9 billion at December 31, 2007, or about 10% of AIG’s total invested assets. • Within AIGI’s $75.3 billion non-agency portfolio, about 89% are AAA-rated and 8% are AA-rated. – Holdings rated BBB or below total approximately $0.5 billion, under 1% of the portfolio and less than 0.1% of total invested assets. – About $7.4 billion (9.9%) of the $75.3 billion is “wrapped” by monoline insurance. – Approximately $2.1 billion of principal was paid down during the fourth quarter. RMBS Type Amortized Cost ($ Billion) % Fair Value ($ Billion) % Agency Pass-Through and CMO Issuances $14.6 16% $ 14.8 17% Prime Non-Agency (incl Foreign and Jumbo MBS related securities) 21.6 24% 21.1 25% Alt-A RMBS 25.3 28% 23.8 28% Subprime RMBS 24.1 27% 21.2 25% Other Housing-Related Paper 4.3 5% 3.9 5% Total RMBS $89.9 100% $84.8 100%
58 AIG Insurance Investment Portfolios RMBS Portfolio Amortized Cost ($ Millions) RATING HOLDINGS AGENCY AAA AA A BBB NON INV TOTAL AGENCY $14,575 $ - $ - $ - $ - $ - $14,575 PRIME JUMBO - 12,708 1,794 483 143 1 15,129 ALT-A - 23,967 994 309 71 8 25,349 SUBPRIME - 20,843 2,833 388 10 - 24,074 SECOND-LIEN - 2,055 40 85 40 3 2,223 HELOC - 1,989 - - - - 1,989 FOREIGN MBS - 5,735 174 294 220 - 6,423 OTHER - 35 13 27 14 - 89 TOTAL $14,575 $67,332 $5,848 $1,586 $498 $ 12 $89,851
59 AIG Insurance Investment Portfolios Total RMBS - $89.9 Billion AAA trancheAAA tranche AA trancheAA tranche A trancheA tranche BBB trancheBBB tranche AA $5.9 Billion (6.5%) A $1.5 Billion (1.7%) BBB $0.5 Billion (0.6%) Equity <$0.1 Billion (0.1%) AAA & Agency $81.9 Billion (91.1%) RMBS (Collateral pool of residential mortgages) RMBS (Collateral pool of residential mortgages) BB and lower Equity tranche BB and lower Equity tranche Payment Waterfall (principal + interest) Priority First Last
60 AIG Insurance Investment Portfolios Total RMBS Exposure by Vintage - $89.9 Billion $Billions 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 AAA 3.95 6.26 7.14 16.75 27.53 20.27 AA 0.07 0.84 0.62 1.24 2.52 0.56 A 0.03 0.26 0.36 0.52 0.28 0.07 BBB 0.00 0.06 0.08 0.22 0.11 0.11 Prior 2003 2004 2005 2006 2007 AIGI focuses almost exclusively on AAA and AA rated investments with relatively short tenors Vintage Weighted average expected life (WAL) is 5.09 years
61 AIG Insurance Investment Portfolios Subprime RMBS • Through its initial investment analysis, combined with ongoing monitoring, virtually all of AIGI’s 2006/2007 exposure is high up in the capital structure (AAAs and AAs). • Nevertheless, lifetime loss estimates for subprime securities – currently 15% - 21% for the 2006 vintage – have been increasing. • Generic AA and AAA rated securities can withstand cumulative loss percentages ranging from the high teens to the mid-to-high 20s. In effect, AIGI is exposed to significant downgrade risk and market price volatility, but AIGI believes that the risk of ultimate loss is not expected to be significant. • Recent Fed rate cuts have been a modest loss mitigant: – The decline in short-term rates increases excess interest, which is used to absorb losses. – Payment shock for ARM borrowers at interest re-set has been significantly reduced. 2006 Vintage Credit Enhancement for AIGI* Rating Original Credit Enhancement Current Credit Enhancement AAA 20.7% 29.6% AA+ and lower 15.5% 21.5% *Source: Intex
62 AIG Insurance Investment Portfolios Subprime RMBS - $24.1 Billion AAA trancheAAA tranche AA trancheAA tranche A trancheA tranche BBB trancheBBB tranche AA $2.8 Billion (11.6%) A $0.4 Billion (1.7%) BBB $10 Million (0.0%) Equity <$0.1 Million (0.0%) AAA $20.9 Billion (86.7%) RMBS (Collateral pool of residential mortgages) RMBS (Collateral pool of residential mortgages) BB and lower Equity tranche BB and lower Equity tranche Last Payment Waterfall (principal + interest) Priority First
63 AIG Insurance Investment Portfolios Subprime RMBS Exposure by Vintage - $24.1 Billion $Billions AIGI focuses almost exclusively on AAA and AA rated investments with relatively short tenors Vintage Weighted average expected life (WAL) is 4.12 years 0.0 2.0 4.0 6.0 8.0 10.0 12.0 AAA 0.14 0.40 0.60 5.81 9.11 4.79 AA 0.01 0.04 0.13 0.33 1.93 0.40 A 0.01 0.08 0.10 0.13 0.04 0.03 BBB - - - 0.01 - - Prior 2003 2004 2005 2006 2007
64 AIG Insurance Investment Portfolios Alt-A RMBS • In the Alt-A sector, AIGI’s initial investment analysis and credit selection process have resulted in most of our exposure being at the super senior* AAA level, especially in the 2006/2007 vintages. • Over 98% of our Alt-A exposure is rated AAA or AA. • While Alt-A loss expectations have climbed into the 2.25 – 3.25% range,** AIGI’s investment decisions with respect to this portfolio have reduced portfolio risk in the current downturn. • Consequently, AIGI is exposed to limited downgrade risk and loss of investment principal in its Alt-A portfolio. *A super senior AAA is structured with credit enhancement in excess of that required by the rating agencies at the AAA level **Source: Lehman Brothers, January 2008, Global Relative Value Conference ***Source: Intex 2006 Vintage Credit Enhancement for AIGI*** Rating Original Credit Enhancement Current Credit Enhancement AAA 19.0% 21.3% AA+ and lower 4.9% 6.5%
65 AIG Insurance Investment Portfolios ALT-A RMBS -$25.3 Billion AAA trancheAAA tranche AA trancheAA tranche A trancheA tranche BBB trancheBBB tranche AA $1.0 Billion (4.0%) A $0.3 Billion (1.2%) BBB $71 Million (0.3%) Equity $8 Million (0.0%) AAA $23.9 Billion (94.5%) RMBS (Collateral pool of residential mortgages) RMBS (Collateral pool of residential mortgages) Payment Waterfall (principal + interest) Priority First BB and lower Equity tranche BB and lower Equity tranche Last
66 AIG Insurance Investment Portfolios ALT-A RMBS Exposure by Vintage - $25.3 Billion $Billions A A AAAAAA AA AA AAAA AA 0.0 2.0 4.0 6.0 8.0 10.0 12.0 AAA 0.21 0.63 0.98 5.15 10.08 6.92 AA 0.03 0.21 0.16 0.44 0.13 0.01 A - 0.04 0.06 0.15 0.05 0.01 BBB - 0.01 0.02 0.04 0.01 - Prior 2003 2004 2005 2006 2007 AIGI focuses almost exclusively on AAA and AA rated investments with relatively short tenors Vintage Weighted average expected life (WAL) is 4.04 years
67 AIG Insurance Investment Portfolios Prime Jumbo RMBS • The domestic prime jumbo RMBS portfolio totaled $15.1 billion at December 31, 2007. • Delinquencies and defaults have generally been well controlled in the prime jumbo market. • Although AIGI took more credit risk in this sector, the majority of our exposure (96%) continues to be in AAAs and AAs. • Loss expectations have climbed but in general remain below the BBB attachment point (approximately 0.5%). • The weighted average expected life of the portfolio is 6.49 years. 2006 Vintage Credit Enhancement for AIGI* Rating Original Credit Enhancement Current Credit Enhancement AAA 8.5% 9.7% AA+ and lower 1.7% 2.1% *Source: Intex
68 AIG Insurance Investment Portfolios RMBS Rating Agency Actions* *Based on 1st Agency to downgrade or put on watch – If on downgrade list, not included on watch list. Source: Moody’s Investors Service, Standard & Poor’s, and Fitch. **January 1, 2008 through February 25, 2008 • Negative rating actions accelerated in the fourth quarter of 2007 and the first quarter of 2008. • The brunt of these actions was borne by the subprime, Alt-A and 2nd lien sectors. The prime jumbo sector was relatively untouched. • An overwhelming majority of the securities on “Watch List Negative” are rated AAA or AA. • Currently we expect very few of the affected securities to ultimately incur a loss to investment principal. Fourth Quarter 2007 YTD 2008** Number of Securities Value ($ Million) Number of Securitie s Value ($ Million) % of Non- Agency RMBS Portfolio Downgraded 35 $443 128 $3,578 4.8% Watch List Negative 96 $3,396 252 $9,694 12.9% Upgrades 0 $0 4 $6 0.0%
69 AIG Insurance Investment Portfolios CMBS Portfolios
70 AIG Insurance Investment Portfolios CMBS Portfolios Overview • AIGI’s CMBS portfolio demonstrated strong credit performance in 2007: – The upgrade / downgrade ratio is significantly better than for the U.S. CMBS universe; and – Delinquencies in underlying collateral are very low and are only approximately 50% of those for the general U.S. CMBS universe. • Approximately 9% of the CMBS exposure (by amortized cost) comes from the re- securitization of CMBS and commercial real estate (CRE) CDOs. Two-thirds of the loans underlying these securities are seasoned 25 months or more. • Our initial investment analysis, combined with ongoing surveillance, has helped protect the CMBS portfolio from the more aggressive underwriting standards that crept into the commercial loan market in 2007. • The portfolio is performing in accordance with expectations from a credit perspective. – For the year ended December 31, 2007, other-than-temporary impairments totaled $147 million, and were recognized primarily as a result of severity of price declines, not credit events. Net unrealized losses were $919 million, about 3.8% of the total portfolio. – No actual credit-related losses to investment principal have been incurred to date.
71 AIG Insurance Investment Portfolios CMBS Portfolios Description Amortized Cost ($ Million) % CMBS (traditional) $21,003 87.81% ReREMIC/ CRE CDO 2,221 9.29% Agency 347 1.45% Other 347 1.45% TOTAL $23,918 100.00% • AIGI’s CMBS portfolio is predominantly comprised of traditional commercial mortgage backed securities. • The underlying collateral of the portfolio is of high quality with close to 89% rated AAA / AA and approximately 99% investment grade. Ratings N/A 1.0% AA 10.9%A 8.7% AAA 78.0% BBB 1.2% BIG 0.2%
72 AIG Insurance Investment Portfolios CMBS Portfolios Vintage % 2007 22.0% 2006 15.5% 2005 18.3% 2004 15.1% 2003 4.9% 2003 & Older 24.2% 100.0% Top 10 States Amortized Cost ($ Million) % NY $3,640 15.2% CA 3,140 13.1% TX 1,476 6.2% FL 1,322 5.5% VA 773 3.2% IL 723 3.0% NJ 696 2.9% PA 588 2.5% GA 576 2.4% MA 555 2.3% $13,489 56.3% • The CMBS portfolio is well-diversified geographically, with the top ten states representing slightly over 56% of total exposure. • The majority of the portfolio is of older vintages, although about 22% is from the 2007 vintage.
73 AIG Insurance Investment Portfolios CMBS Portfolios – ReREMIC/ CRE CDO Holdings Rating Amortized Cost ($ Million) % AAA/Aaa $1,013 45.6% AA/Aa 229 10.3% A/A 826 37.2% BBB/Baa 131 5.9% BB/Ba 22 1.0% $2,221 100.0% Top 10 States % CA 15.2% NY 10.2% TX 7.2% FL 5.7% VA 3.6% IL 3.6% PA 3.1% NJ 2.9% GA 2.8% MA 2.7% 57.0% • Only 9.3% of the total CMBS portfolio is represented by ReREMIC / CRE CDO securities. – Close to 58% of these securities are ReREMIC and the other 42% are CRE CDOs – Highly rated collateral (99% investment grade rated) – Geographically diversified – Well-seasoned portfolio with over 65% of the underlying loans seasoned over 25 months Collateral Seasoning (mos) 37+ 45.1% </= 12 3.5% 13 – 24 31.1% 25-36 20.3%
74 AIG Insurance Investment Portfolios CMBS Portfolios Current Delinquency % Source: Trepp, Morgan Stanley and Intex Delinquencies as of 2/4/08 • The U.S. CMBS sector has enjoyed strong credit performance. Delinquencies are near historic lows and have remained below 1% since 2005. • AIGI’s CMBS portfolio is currently outperforming the U.S. CMBS universe. 0.04 0.02 0.13 0.12 0.06 0.02 0.27 0.07 0.50 0.23 0.00 0.10 0.20 0.30 0.40 0.50 0.60 60 90+ FCL RE O Total US CM B S Univers e A IG CM B S P ortfolio
75 AIG Insurance Investment Portfolios CMBS Portfolios Rating Actions Full Year 2007 AIGI CMBS Portfolio U.S. CMBS Universe Combined 43.2:1 8.4:1 Investment Grade Bonds 43.2:1 21.4:1 Below Investment Grade Bonds n/a 1.8:1 • AIGI’s CMBS upgrade / downgrade ratio is significantly better than that of the U.S. CMBS universe.
76 AIG Insurance Investment Portfolios CDOs
77 AIG Insurance Investment Portfolios CDO Overview • Investments in collateralized debt obligations (CDOs) total $4.55 billion representing approximately 1% of total insurance investment portfolios. – Only $58 million are mezzanine ABS CDOs with subprime exposure. – Only 0.4% of total CDO holdings were downgraded in 2007. • CDO pricing has been adversely affected by current market conditions. • Based on our current analysis, the risk of ultimate loss of investment principal is low.
78 AIG Insurance Investment Portfolios CDO Overview • As of December 31, 2007 the composition of the $4.55 billion CDO global portfolio is as follows: • The weighted average market price of the CDO portfolio was $81(2) as of December 31, 2007, down from $88(2) as of September 30, 2007. (1) Below Investment Grade (2) As compared to par of $100 Ratings ($ in Billions) Amortized Cost % AAA $0.80 17.6% AA 1.02 22.5% A 2.21 48.5% BBB 0.42 9.2% BIG(1) and Equity 0.10 2.2% Total $4.55 100% Collateral Type ($ in Billions) Amortized Cost % Bank Loans (CLO) $2.14 47.0% Synthetic Investment Grade 1.56 34.3% Other 0.79 17.4% Subprime ABS 0.06 1.3% Total $4.55 100%
79 AIG Insurance Investment Portfolios Bank Loan CLOs • The composition of the $2.14 billion in collateralized loan obligations (CLO) holdings (47% of the total CDO portfolio) by rating is as follows: • Holdings continue to exhibit good performance, reflecting the relatively low default rate environment for leveraged loans. We expect the default rate to increase, but to still moderate levels. • None of the debt tranches in our CLO portfolio has been downgraded or placed on negative watch by the rating agencies. • None of our tranches is deferring interest. • 80% of CLO holdings are rated A or better. Ratings Amortized ($ in Billions) Cost % AAA $0.02 1% AA 0.13 6% A 1.55 73% BBB 0.35 16% BIG and Equity 0.09 4% Total $2.14 100%
80 AIG Insurance Investment Portfolios Synthetic Investment Grade Rated Corporate CDOs • The $1.56 billion in Investment Grade Corporate Synthetic holdings (34% of the total CDO portfolio) is highly rated as shown below: • The underlying portfolios are predominantly CDS against investment grade corporate bonds and loans. • Holdings continue to exhibit good performance, reflecting the relatively benign default environment for investment grade rated corporates. The initial credit enhancement levels remain intact. • None of our tranches has been downgraded or placed on negative watch by the rating agencies. • None of our tranches is deferring interest. • 76% of these holdings are rated AA or better and 99% are rated A or better. Ratings ($ in Billions) Amortized Cost % AAA $0.44 28% AA 0.74 48% A 0.37 23% BBB 0.01 1% Total $1.56 100%
81 • 91% of the remaining CDO holdings, which include primarily market value and older vintage CDOs, is rated A or better. • These holdings of $793 million have maintained their value ($785 million fair value). • These CDO holdings are experiencing stable performance. • The composition of the holdings from a ratings standpoint is as follows: AIG Insurance Investment Portfolios Other CDOs BIG and Equity 2% BBB 7% AAA 41% AA 17% A 33%
82 AIG Insurance Investment Portfolios Monoline Exposure
83 AIG Insurance Investment Portfolios Monoline Exposure • AIGI’s exposure to monolines totals $42.2 billion, 99% of which are financial guarantees. – 74% of the total exposure relates to municipal bonds, which are highly rated, even without the financial guarantees. – AIGI does not rely primarily on financial guarantees with regards to making investment decisions in its bond portfolio. • The composition by asset class is as follows: 1) Refers to cash collateral accounts in certain synthetic CDOs. $399 million of this exposure is investment agreements with financial guarantee insurance policies provided by the monolines (includes $220 million of fully collateralized investment agreements and $179 million of investment agreements which are subject to collateral posting requirements, should the monoline guarantor be downgraded). Also includes $41 million in an investment agreement issued by a monoline with a corporate guarantee provided by a highly rated non-monoline guarantor. (2) Represents amortized cost and fair value related to $57 million of bonds and $136 million notional of CDS. (3) The fair value for the bond portion is $52 million and the market value for the CDS portion is ($13) million. Asset Class Amortized Fair ($ in Billions) Cost Value Municipals $31.41 $31.88 RMBS/CMBS 7.40 6.82 ABS 2.14 2.01 Corporates 0.76 0.80 Investment Agreements in CDOs (1) 0.44 0.36 Total Insured 42.15 41.87 Direct Corporate Exposure (2),(3) 0.06 0.04 Total Exposure $42.21 $41.91
84 AIG Insurance Investment Portfolios Monoline Exposure • Exposure by monoline entity and by type of exposure is as follows: (1) Amounts above are exclusive of $136 million in Notional of CDS as follows: $56 million (AMBAC), $29 million (MBIA), and $51 million (Assured Guaranty). (2) Other includes the amortized cost of corporate exposure and Investment Agreements in CDOs. Financial Guarantees Amortized Cost ($ in Billions) (1) Other (2) MBIA $13.10 $0.16 FSA 10.45 0.14 AMBAC 9.13 0.15 FGIC 7.37 0.04 SCA (XLCA/XLFA) 0.76 - CIFG 0.67 - Assured Guaranty Corp. 0.22 - Multiple 0.01 - $41.71 $0.49
85 AIG Insurance Investment Portfolios Monoline Exposure by Underlying Ratings • The quality of the insured portfolio is high, with 84% having ratings of A or above. This is also a reflection of the large percentage of municipals in our wrapped monoline portfolio. (1) Includes RMBS/CMBS and ABS underlying ratings, which are based solely on AIG’s internal ratings assessment. (2) Excludes $440 million of Investment Agreements in CDOs and $57 million of corporate exposure. Underlying Ratings (1) Amortized Cost ($ in Billions) (2) % of Total AAA $2.90 7.0% AA 21.35 51.1% A 10.89 26.1% BBB 3.62 8.7% BB 2.40 5.8% B 0.04 0.1% CCC 0.45 1.1% Non Rated 0.06 0.1% $41.71 100.0%
86 AIG Insurance Investment Portfolios Monoline Exposure Municipal Bonds by Underlying Ratings • As of December 31, 2007 AIGI’s municipal bond portfolio totaled $61.5 billion. • More than 99% of the total municipal bond portfolio is rated A or better, even without the financial guarantees. • More than 98% of AIGI’s insured municipal bond portfolio is rated A or better, even without the financial guarantees. Insured Portfolio ($31.4 Billion) No Underlying 0.1% Baa/BBB 1.2% AAA .4% Aa/AA 65.9% A 32.4% Total Portfolio ($61.5 Billion) Aa/AA 53.1% Aaa/AAA 14.6% AAA Pre-re/ETM 12.2%A 19.4% No Underlying 0.1% Baa/BBB 0.7%
87 AIG Insurance Investment Portfolios Monoline Exposure RMBS/CMBS • AIGI’s insured RMBS/CMBS portfolio is diversified across product types. • 63% of the underlying RMBS/CMBS portfolio is internally rated investment grade. • Below Investment Grade represents primarily second lien and HELOC pools that have experienced worse than anticipated performance. AIG Internal Ratings Amortized ($ in Billions) Cost % AAA $2.58 35% AA 0.32 4% A 0.42 6% BBB 1.31 18% BB 2.32 31% CCC 0.45 6% $7.40 100% Asset Class Amortized ($ in Billions) Cost % SECOND LIEN $1.96 26% HELOC 1.93 26% ALT-A 1.69 23% Subprime 1.26 17% JUMBO 0.42 6% Foreign MBS 0.06 1% CMBS 0.05 1% Manufactured Housing 0.03 0% $7.40 100%
88 AIG Insurance Investment Portfolios Monoline Exposure ABS • AIGI’s insured non-RMBS/CMBS ABS portfolio is diversified across product types. • 98% of the underlying ABS portfolio is internally rated investment grade. • We do not rely on monoline support as the primary source of repayment and thus do not look to it as a material consideration to the collectibility of the portfolio. Asset Class Amortized ($ in Billions) Cost % Business/Franchise Loan $0.55 26% Auto Loan 0.48 22% Future Flow 0.42 19% Lot Loan 0.26 12% Project Finance & Other 0.19 9% Railcar Loan/Lease 0.10 5% Timeshare 0.10 5% Credit Card 0.04 2% $2.14 100% AIG Internal Ratings Amortized ( $ in Billions) Cost % AAA $0.20 9% AA 0.16 7% A 0.25 12% BBB 1.49 70% BB 0.04 2% CCC 0.00 0% $2.14 100%
89 AIG Insurance Investment Portfolios Monoline Exposure • Sound fundamental credit underwriting is the foundation for all of AIGI’s investment decisions, and financial guarantees are viewed as a secondary form of payment for all municipal and other wrapped investments. – More than 99% of the underlying $31.4 billion insured municipal bond portfolio is rated investment grade even without the financial guarantees. – 63% of the underlying $7.4 billion insured RMBS/CMBS portfolio is internally rated investment grade. – 98% of the underlying $2.1 billion insured ABS portfolio is internally rated investment grade. • Currently there are 10 RMBS Second Lien, Home Equity and Subprime transactions totaling $380 million, or less than 1% of AIGI’s total insured portfolio, that are known to be receiving contractual payments through their financial guarantees.
90 AIG Insurance Investment Portfolios: Conclusion
91 AIG Insurance Investment Portfolios In Conclusion • The global financial markets remain under considerable stress with reduced financing opportunities for residential mortgage borrowers. • Market expectations for losses in the 2006/2007 U.S. subprime and Alt-A vintages have increased substantially. • Our preference for exposures higher in the capital structure limits our risk of investment principal loss, yet our subprime holdings remain exposed to material market price volatility and downgrade risk. • Our Alt-A holdings continue to benefit from our decision to require credit-enhanced levels in deals beyond the AAA enhancement required by the rating agencies. • Mark-to-market write-downs have increased, but ultimate investment principal losses are not expected to be significant. • Though our CMBS holdings have suffered from volatile market pricing, underlying fundamentals remain strong, with very low delinquencies and a high upgrade / downgrade ratio. • Our CDO holdings have suffered minimal downgrades in 2007, and we have little exposure to the currently struggling subprime CDO market. • Excluding municipal bonds, AIGI’s exposure to monoline insurers is a moderate percentage of our bond holdings. None of these holdings relied on financial guarantees as a primary source of repayment at the time of acquisition. • We view monoline insurance as a secondary consideration to the credit-worthiness of the municipal bond portfolio.
Mortgage Guaranty
93 United Guaranty (UGC) • UGC, as a broad market participant, operates in an inherently cyclical business that is highly correlated to the fortunes of the housing market. • UGC prices for long-term profitability to absorb market disruptions and has cumulatively generated $2.6 billion in pre-tax operating income and maintained a cumulative loss ratio of 50% for the 10-year period ended December 31, 2007. • The quality of new business production is improving, driven by UGC’s underwriting and eligibility adjustments, along with more rigorous underwriting standards applied by UGC’s lender customers. The positive effects of these changes will be reflected in future years’ results. • UGC is well positioned to take advantage of the opportunities presented when the market emerges from this housing correction. Executive Statement
94 United Guaranty • Mortgage Guaranty is a multi-year contract with monthly premiums and automatic renewals (15-30 year mortgage term). UGC can generally only cancel the policy for non-payment of premium or other policy exclusions. • Mortgage Guaranty first-lien price changes are slow to affect results, as they are regulated by state Departments of Insurance and require changes to loan origination systems by large mortgage lenders. • Mortgage Guaranty performance is predominantly determined by macroeconomic events in the early years of the policy. Current year loss expenses are driven by loans from prior vintage years. • This business model results in a portfolio with an average life of 5-7 years, with new production contributing less than 20% of the calendar year net premiums written but building a base for renewal premiums. Mortgage Guaranty Product Characteristics Changes in underwriting and eligibility guidelines positively affect future results, but do not have a significant effect on current year results.
95 United Guaranty December 31, 2007 This table is for informational purposes only. Net Risk-in-Force (RIF) = Insurance risk on mortgages net of risk sharing and reinsurance. Loans with unknown FICO scores are included in the FICO (620-659) based on similar performance characteristics. Delinquency figures are based on number of policies (not dollar amounts), consistent with mortgage insurance industry practice. Real Estate Portfolio Total Portfolio FICO (≥ 660) FICO (620-659) FICO (<620) Domestic Mortgage Net Risk-in-Force 60+ Day Delinquency $29.8 Billion 3.71% $21.0 Billion 2.10% $6.3 Billion 6.60% $2.5 Billion 15.56% 2007 Vintage 60+ Day Delinquency $8.9 Billion 2.46% $6.2 Billion 1.07% $1.9 Billion 3.96% $886 Million 12.99% 2006 Vintage 60+ Day Delinquency $6.5 Billion 4.47% $4.5 Billion 2.65% $1.4 Billion 7.68% $670 Million 18.12% 2005 Vintage 60+ Day Delinquency $5.1 Billion 3.69% $3.7 Billion 2.39% $1.0 Billion 7.10% $292 Million 15.17% 2004 Vintage 60+ Day Delinquency $3.3 Billion 3.36% $2.4 Billion 1.73% $701 Million 6.81% $214 Million 16.80% Loans > 95% 60+ Day Delinquency $10.4 Billion 4.32% $6.6 Billion 2.13% $2.6 Billion 7.38% $1.1 Billion 16.44% Low Documentation 60+ Day Delinquency $5.6 Billion 3.89% $5.0 Billion 3.25% $532 Million 8.10% $112 Million 18.29% Interest Only & Option ARMs 60+ Day Delinquency $2.9 Billion 8.82% $2.4 Billion 7.76% $433 Million 13.07% $81 Million 16.68%
96 United Guaranty • The deterioration of the U.S. housing market has affected all segments of the mortgage business, but the high LTV second-lien product is particularly sensitive to declining home values and as a result constitutes a disproportionate share of 2007 losses. • Due to the accelerated claims cycle for second-lien mortgages, these net losses incurred should work through the portfolio much faster and peaked in 2007. • However, first-lien net losses incurred are beginning to negatively impact operating results as delinquencies progress through the claim cycle. Further deterioration is anticipated in 2008. • As of 12/31/07, expected future losses are significantly below net risk-in-force. Future premiums are expected to exceed future losses on the existing portfolio. Domestic First Lien - $697MM 49% of losses incurred Domestic Second Lien - $737MM 51% of losses incurred Domestic Second Lien - $3.8BN 13% of portfolio Domestic First Lien - $26.0BN 87% of portfolio United Guaranty Domestic Mortgage Net Risk-In-Force December 31, 2007 Although quality of new business production is improving, near-term results will continue to reflect market downturn. Loss Emergence Domestic Second Lien - $3.8BN 13% of portfolio Domestic First Lien - $26.0BN 87% of portfolio United Guaranty Domestic Mortgage Net Risk-In-Force December 31, 2007 Domestic First Lien - $697MM 49% of net losses incurred Domestic Second Lien - $737MM 51% of net losses incurred United Guaranty Domestic Mortgage Net Losses Incurred Total Year 2007
97 United Guaranty Analysis of Loss Reserve – Domestic Mortgage Product • UGC’s Corporate Actuarial Department employs rigorous analyses of the loss reserve adequacy of its businesses on a quarterly basis - The total loss reserve equals the sum of the case reserves and incurred but not reported (IBNR) reserves. • In the actuarial testing of loss reserve adequacy, a variety of data and methods are employed - Accident year data is the primary focus, which represents the date of first missed payment on a loan. - Reserving methods typically include: paid development, incurred development, Cape Cod, Bornhuetter-Ferguson and incurred count severity. - A range of reserve estimates is established based on observed historical variance in loss reserve estimates and a selected confidence level. - An updated analysis of the case reserve and IBNR factors is performed on a quarterly basis. • The actuarial analysis results, together with any recommended changes in reserves, are reviewed on a quarterly basis and approved by UGC’s CFO, Controller and Chief Risk Officer, as well as by AIG’s Chief Actuary, Chief Credit Officer and the CFO of AIG’s Domestic Brokerage Group. Mortgage Guaranty Insurance accounting requires reserves to be established based upon current delinquencies, but does not permit any provision for future delinquencies.
98 • $26.0 billion net risk-in-force • 878,829 policies in force • Average FICO score of 696 • 48,263 delinquent loans • 5.49% delinquency ratio United Guaranty First-Lien Portfolio In-Force Summary as of December 31, 2007
99 4.94 5.01 4.69 4.41 4.26 4.29 4.29 4.39 4.49 4.51 4.59 4.62 4.68 4.92 4.73 4.52 4.44 4.52 4.68 4.89 5.06 5.28 5.65 5.93 6.33 3.70 3.76 3.51 3.26 3.14 3.20 3.26 3.36 3.39 3.48 3.56 3.59 3.72 3.91 3.74 3.56 3.56 3.71 3.98 4.23 4.39 4.65 4.89 5.30 5.69 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00 Dec- 05 Jan- 06 Feb- 06 Mar- 06 Apr- 06 May- 06 Jun- 06 Jul- 06 Aug- 06 Sep- 06 Oct- 06 Nov- 06 Dec- 06 Jan- 07 Feb- 07 Mar- 07 Apr- 07 May- 07 Jun- 07 Jul- 07 Aug- 07 Sep- 07 Oct- 07 Nov- 07 Dec- 07 Industry (excluding UGC, Radian) Domestic First-Lien United Guaranty Delinquency Rates – UGC vs. Industry (First-Lien Primary) Industry* United Guaranty The first-lien mortgage delinquency ratio has consistently run below the industry average, although the gap is narrower than historical levels. *Source: Mortgage Insurance Companies of America (MICA) Figures (for UGC and industry) are based on primary insurance and do not include pool insurance. %
100 First-Lien Portfolio by Product 78% of the first-lien portfolio is in fixed rate loans. Portfolio Net RIF by Product as of December 31, 2007 $1.1BN 4% $20.2BN 78% $2.9BN 11% $1.8BN 7% Fixed Rate Loans Positively Amortizing ARM’s Potential Neg-Am ARM’s Interest Only Loans United Guaranty
101 First-Lien Portfolio by Flow / Bulk Channel UGC chose to be a minor participant in the high risk bulk channel. Portfolio Net RIF by Channel as of December 31, 2007 $1.2BN 5% $24.8BN 95% Flow Channel Bulk Channel United Guaranty
102 First-Lien Key Risk Initiatives UGC continues to implement key risk initiatives to improve the quality of new business production. • Tightened underwriting standards and eligibility guidelines in conjunction with market movement. • Increased rates in select, higher risk business segments. • Established new and modified existing portfolio concentration caps. • Flight to quality: – Improved mortgage insurance penetration (fewer “piggybacks”). – Increased conforming (GSE eligible) loan production. – Improved quality of new business production. United Guaranty
103 • $ 3.8 billion net risk-in-force • 643,147 policies in force • Average FICO score of 716 • 8,205 delinquent loans • 1.28% delinquency ratio In-Force Summary as of December 31, 2007 United Guaranty Second-Lien Portfolio
104 88% of the second-lien portfolio is in high credit quality loans. Portfolio Net RIF by FICO score as of December 31, 2007 $3.4BN 88% $0.4BN 11% FICO ≥ 660 FICO 620 – 659 FICO < 620 $59MM 1% United Guaranty Second-Lien Portfolio by FICO Score
105 • Strengthened Underwriting Guidelines – Eliminated Alternative Risk product in fourth quarter 2006. – Eliminating 100% Combined LTV (CLTV) purchase money (“piggyback”) seconds. – Eliminated significant segments of stated income and third party originated business. • Reduced Risk Retention Levels – In lieu of 100% coverage, introducing co-insurance to align the lenders’ interests with those of UGC. – Utilizing mezzanine risk layers and lower policy limits (policy limit is commonly referred to as stop loss in the mortgage insurance industry). • Improved Pricing – Implementing higher pricing on new business. – Utilizing experience and retrospective rating plans more frequently. • Enhanced Portfolio Risk Management – Established several new portfolio concentration caps in addition to modifying selected existing caps. UGC continues to implement key risk initiatives to improve the quality of new business production. United Guaranty Second-Lien Key Risk Initiatives
106 United Guaranty Risk Mitigating Factors • UGC uses several mitigants to reduce the volatility of losses transferred from lenders, which are reflected in the net risk-in-force figures: - Risk sharing: funded arrangements through captive reinsurance with most major lenders, in which the lenders share in losses above a determined attachment point. - Reinsurance: quota share reinsurance on a portion of UGC’s sub-prime first-lien product and segments of its second-lien product. - Policy limit: second-lien mortgage business has an aggregate policy limit provision limiting losses to a percentage (generally 10%) of the total original loan balances in each policy. - Fraud exclusion: UGC maintains a fraud exclusion on both its first-lien and its second-lien mortgage businesses. • 78% of first-lien mortgages are fixed rate, which have about 50% lower delinquency than ARMs. • First-lien mortgages consist of 87% single family residences and 93% owner occupied.
107 United Guaranty • UGC is engaged in a highly cyclical business, with high returns in 8 out of 10 years and underwriting losses in 2 out of 10 years on average. The downturns in the housing industry negatively affect short-term profitability, as pricing is actuarially derived for the long-term performance. • UGC expects that the downward market cycle will continue to adversely affect its operating results for the foreseeable future and is likely to result in another significant operating loss in 2008. • UGC has re-engineered its second-lien product, further tightened first-lien eligibility guidelines and increased rates in select, high-risk business segments. • The quality of new business production is improving, driven by UGC’s underwriting and eligibility adjustments, along with more rigorous underwriting standards applied by UGC’s lender customers. The positive effects of these changes will be reflected in future years’ results. • Additional positive market trends include: • Improved mortgage insurance market penetration. • Increased conforming (GSE eligible) loan products. • Improved persistency of insured portfolio. Summary
Consumer Finance
109 American General Finance • For over 50 years, AIG’s domestic consumer finance business has provided mortgage and consumer loans through a network of branch offices, which currently consists of 1,600 locations. • In addition, AGF originates and acquires mortgage loans through its centralized mortgage operations.
110 $0 $5 $10 $15 $20 $25 $30 YE04 YE05 YE06 YE07 Real Estate Non-Real Estate Retail Sales Finance $ Billions $20.2 7% 15% 6% 14% $23.8 80% 8% 14% 78% $24.3 $25.5 9% 15% 76% 78% Total Receivables Before Allowance American General Finance Portfolio Mix
111 American General Finance December 31, 2007 This table is for informational purposes only. AGF’s loan underwriting process does not use FICO scores as a primary determinant for credit decisions. AGF uses proprietary risk scoring models in making credit decisions. Delinquency figures are shown as a percentage of outstanding loan balances, consistent with mortgage lending practice. Differences in totals by columns and rows are due to rounding. Outstandings LTV 80% 84% 80% 75% 60+% 2.64% 1.29% 3.05% 4.48% 2007 Vintage LTV 77% 82% 79% 74% 60+% 0.99% 0.44% 1.01% 1.33% 2006 Vintage LTV 80% 87% 81% 75% 60+% 3.57% 1.70% 3.02% 5.26% 2005 Vintage LTV 82% 85% 82% 76% 60+% 3.03% 1.65% 4.49% 5.60% 2004 Vintage LTV 82% 83% 80% 75% 60+% 1.98% 1.12% 3.71% 6.06% LTV Greater than 95.5% LTV 99% 99% 99% 98% 60+% 2.36% 1.81% 5.10% 5.69% Low Documentation LTV 76% 78% 76% 71% 60+% 4.30% 3.86% 4.22% 5.97% Interest Only LTV 89% 89% 88% 79% 60+% 3.12% 2.55% 5.12% 11.90% Total Portfolio FICO (≥ 660) FICO (620 - 659) FICO (< 620) $19.5 Billion $9.7 Billion $3.3 Billion $6.2 Billion $4.0 Billion $1.3 Billion $830.7 Million $1.9 Billion $3.4 Billion $1.2 Billion $647.6 Million $1.5 Billion $4.8 Billion $2.9 Billion $856.0 Million $1.1 Billion $4.5 Billion $3.5 Billion $554.0 Million $503.1 Million $3.5 Billion $3.0 Billion $385.7 Million $169.4 Million $531.9 Million $283.5 Million $165.4 Million $83.0 Million $1.7 Billion $1.4 Billion $284.3 Million $23.8 Million
112 American General Finance Net Real Estate Loan Growth $ Billions $1.4 $1.2 $0.4 $0.3 $0.1 $0.1 -$0.1 -$0.3 $0.0 $0.2 $0.3 $0.0 -$0.6 $0.0 $0.6 $1.2 $1.8 1Q05 2Q05 3Q05 4Q05 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 As the real estate market softened, AGF maintained its underwriting discipline despite experiencing lower volume and growth.
113 0% 5% 10% 15% 20% 25% 2004 2005 2006 1Q07 2Q07 3Q07 4Q07 AGF Total Real Estate 60+ Day Delinquency*60+ Day Delinquency* * Source: Company and industry reports Subprime ABS Real Estate MarketSubprime ABS Real Estate Market American General Finance AGF vs. “Subprime ABS Market”
114 0% 1% 2% 3% 4% 5% YE03 YE04 YE05 YE06 YE07 American General Finance Real Estate Credit Quality AGF’s delinquency and losses have risen from recent all- time lows, yet remain below their targets as well as the industry during the current mortgage market conditions. Net Charge-off Target .75% - 1.25% 60+ Day Delinquency Target 3.0% - 4.0% Target ranges were set by AGF management years ago to denote sound credit quality parameters.
115 American General Finance Risk Mitigating Practices - Real Estate Portfolio • 97% of mortgages are underwritten with full income verification. • 88% are fixed-rate mortgages; only about 9% of the total mortgage portfolio re-sets interest rates by the end of 2008; about 10% by the end of 2009. • Adjustable rate mortgages (ARMs): borrowers are qualified on a fully-indexed and fully-amortizing basis at origination. • No delegation of underwriting to unrelated parties. • No Option ARMs. • Substantially all loans are: – First mortgages (92%) – Owner occupant borrowers (94%) • Geographically diverse portfolio.
116 • AGF’s allowance for loan losses is maintained at a level considered adequate to absorb management’s best estimate of credit losses in the existing portfolio. • AGF’s Credit Strategy and Policy Committee is responsible for determining the appropriate level for the allowance. – Membership consists of AGF’s senior management, including, among others, AGF’s CEO, the Executive Vice President of AGF’s Branch Operations, AGF’s CFO and AGF’s Chief Risk Officer – The Committee evaluates both internal and external factors including: • The composition of AGF’s finance receivable portfolio • Prior finance receivables losses and delinquency experience • Results of migration and Monte Carlo analyses • Current economic environment • AGF calculates three different migration scenarios based on varying assumptions to evaluate a range of possible outcomes for the quantitative component of the allowance for residential real estate. • Conclusions reached by the Committee are reviewed on a quarterly basis and approved by AIG’s Chief Credit Officer and the CFO of AIG’s Financial Services Division. American General Finance Allowance Methodology
117 0% 1% 2% 3% 4% 5% 2004 2005 2006 2007 0% 1% 2% 3% 4% 5% Charge-Offs Allowance for Loan Loss Total PortfolioTotal Portfolio 1.6X 2.0X 2.1X Coverage Ratio ($ Allowance / $ Charge-offs) 2.1X American General Finance Strong Credit Quality
118 American General Finance Equity One Portfolio Acquisition • AGF has agreed to acquire $1.5 billion in outstanding balances of branch- based consumer loans of Equity One, Incorporated. • Equity One’s branch-offered products are similar to those of AGF (1st & 2nd Fixed Rate Mortgages, Consumer Loans, Retail Sales). • Customer profile and credit quality performance are also very similar to those of AGF. • The transaction is expected to close during the first quarter of 2008. • Acquisition of 130,000 new accounts with minimal overhead increase. • AGF was able to acquire the portfolio at an attractive price given current market conditions. • Acquisition is expected to be accretive to earnings in 2008.
119 American General Finance Summary • At the end of the fourth quarter, AGF’s real estate loan portfolio remained at $19.5 billion, flat to the end of the third quarter. – The 2007 vintage production is the result of balanced growth from both its centralized real estate and branch operations which met both strict underwriting guidelines and return hurdles in a challenging market. • AGF maintained its time-tested, disciplined underwriting approach throughout the residential real estate boom, continually re-evaluating guidelines and adjusting as appropriate, resulting in: – Lower volume – Better than targeted delinquency and charge-off – Better than industry-experienced delinquency and charge-off • AGF believes that the housing market will likely continue to deteriorate for the remainder of 2008, but the company’s business model and strict underwriting approach are sound, allowing the company to pursue opportunities.
AIG Conference Call Credit Presentation - Supplemental Materials - February 29, 2008
Capital Markets – Data Appendices
A-2 Summary StatisticsSummary Statistics ““Super SeniorSuper Senior”” Credit Derivatives*Credit Derivatives* Transaction Type Corporate – Regulatory Capital Motivated European Mortgage Corporate - Arbitrage Motivated Multi-Sector CDOs Transactions w/Mixed Transactions Collateral, including w/No Sub Prime Sub Prime Collateral Gross Notional ($ Billion) 306.0 182.8 87.3 82.8 27.3 AIGFP Net Notional Exposure ($ Billion) 229.6 149.1 70.4 61.4 16.8 Number of Transactions 58 35 36 103 13 Weighted Average Subordination (%) ¹ 22.0% 13.8% 18.3% 23.3% 18.0% Average Number of Obligors / Transaction 1,571 74,819 122 194 185 Expected Maturity (Years) 1.2 ² 2.3 ² 4.0 5.0 ³ 5.4 ³ 1. Weighted by Gross Transaction Notional 2. Maturity shown reflects first non regulatory call date, although majority of transactions have Regulatory Capital Calls from Jan 08 3. Reflects the Weighted Average Life *All data is as of December 31, 2007.
A-3 Summary of Key Statistics for Multi SectorSummary of Key Statistics for Multi Sector CDOsCDOs withwith Mixed Collateral, Including Sub PrimeMixed Collateral, Including Sub Prime (Billion US$)(Billion US$) • Total Gross Transaction Notional: 53.1 • AIGFP Total Net* Notional: 43.5 • Number of Transactions: 45 • Average Attachment: 15.4% Sub Prime Reference Obligations • Gross Sub Prime Notional: 25.2 • AIGFP Net* Sub Prime Notional: 17.3 • Average Sub Prime: 47.5% • Avg. HPA for Sub Prime Collateral: 7.7% • Average Cum. Loss Rate: 1.2% • Average FICO: 625 • Total Gross Transaction Notional: 29.7 • AIGFP Total Net* Notional: 17.9 • Number of Transactions: 58 • Average Attachment: 37.3% Sub Prime Reference Obligations • Gross Sub Prime Notional: 18.9 • AIGFP Net* Sub Prime Notional: 8.2 • Average Sub Prime: 63.6% • Avg. HPA for Sub Prime Collateral: 7.6% • Average Cum. Loss Rate: 1.1% • Average FICO: 624 High Grade Collateral (predominantly AA Rated) Mezzanine Collateral (predominantly BBB Rated) * Net of all Transaction subordination Information shown is sourced from LoanPerformance except in the following circumstances: • Underlying Reference Obligation data is sourced from Intex • Loss Data is sourced from CSFB’s “Locus” system • Sector Categorization is compiled from Individual Transaction Trustee Reports and Moody’s and S&P data
A-4 ““Super SeniorSuper Senior”” ExposureExposure Corporate LoansCorporate Loans
A-5 Regulatory CapitalRegulatory Capital -- Corporate by Primary JurisdictionCorporate by Primary Jurisdiction Net Notional (Bn) Percentage of Total Current Average Subordination Realised Pool Losses to Date % Weighted Average Maturity (years) Number of Deals Primarily Single Country Exposure Portfolio To First Call * To Maturity Germany 19.9 8.7 24.7 0.09 2.2 8.9 10 USA 7.1 3.1 40.0 0.00 0.8 12.3 1 Netherlands 4.7 2.0 18.2 0.00 2.0 46.0 1 Portugal 4.2 1.8 11.6 0.08 0.8 11.8 1 UK 2.2 1.0 25.0 0.00 1.0 13.8 1 France 2.1 0.9 21.4 0.00 1.0 1.0 1 Australia/New Zealand 1.8 0.8 9.0 0.00 1.7 3.2 1 Finland 1.2 0.5 18.0 0.00 1.0 7.0 1 Belgium 3.7 1.6 28.4 0.05 1.5 6.1 2 46.9 Regional Exposure Portfolio USA Majority 66.5 29.0 25.6 0.00 1.4 5.9 15 W. Europe Majority 104.5 45.5 18.3 0.00 0.9 8.6 16 Asia/Australia Majority 10.7 4.6 16.8 0.00 0.7 3.1 6 Emerging Market 1.0 0.5 29.0 0.00 3.3 3.3 2 182.7 Total 229.6 100.0 22.0 0.01 1.2 8.4 58 * The vast majority of deals have regulatory calls from January 2008. We expect that these calls will be exercised over the next 12-18 months as the different originating banks in Europe are able to adopt the new Basle II Capital standards. The call date listed in the chart is the first non regulatory call.
A-6 Arbitrage MotivatedArbitrage Motivated -- Corporate by Primary JurisdictionCorporate by Primary Jurisdiction Net Notional (Bn) Percentage of Total Current Average Subordination Realised Pool Losses to Date % Weighted Average Maturity (years) Number of Deals Primarily Single Country Exposure Portfolio To First Call To Maturity USA 50.2 71.3 16.8 0.08 4.7 4.7 23 Regional Exposure Portfolio USA Majority 18.1 25.7 15.7 0.69 1.7 1.7 7 W. Europe Majority 2.1 3.0 49.2 0.00 5.0 14.1 6 20.2 Total 70.4 100.0 18.3 0.22 4.0 4.5 36
A-7 ““Super SeniorSuper Senior”” ExposureExposure European Residential MortgagesEuropean Residential Mortgages
A-8 European Residential Mortgages Summary by GeographyEuropean Residential Mortgages Summary by Geography AIGFP Net Notional Exposure (Bn) % of Total Exposure Current Average Subordination % Realised Losses to Date % of Pool Weighted Average Maturity (years) Number of Transactions To First Call* To Maturity Denmark 38.9 26.2 9.2 0.00 1.4 31.8 3 France 38.0 25.5 8.2 0.01 2.0 31.6 7 Germany 36.3 24.3 19.0 0.10 1.7 42.1 17 Netherlands 22.8 15.3 21.7 0.01 3.8 10.2 4 Sweden 9.6 6.4 12.9 0.00 2.0 34.5 2 UK 1.8 1.2 10.0 0.00 1.2 31.2 1 Spain 1.7 1.1 9.2 0.00 9.5 42.1 1 Total 149.1 100.0 13.8 0.03 2.3 30.3 35 * All of these deals have regulatory calls from January 2008. We expect that these calls will be exercised over the next 12-18 months as the different originating banks in Europe are able to adopt the new Basle II Capital standards. The call date listed in the chart is the first non regulatory call.
A-9 ““Super SeniorSuper Senior”” Multi Sector CDO ExposureMulti Sector CDO Exposure Consisting of Mixed Collateral,Consisting of Mixed Collateral, including Sub Prime:including Sub Prime: Mezzanine Collateral Underlying SummaryMezzanine Collateral Underlying Summary
A-10 Summary of Transaction Current StatusSummary of Transaction Current Status Gross Notional 29.7bn AIG Notional Exposure (net of transaction subordination) 17.9bn Max: 904 mm Min: 31 mm Avg: 309 mm Number of Transactions 58 Managed Transactions 21 Static Transactions 37 Average Number of Obligors 172 Average AIG Attachment Point (weighted by transaction notional) 37.3% Max: 74.8% Min: 24.5% Average % of AIG Subordination that is AAA Rated by at least one agency 36.7% Max: 62.3% Min: 7.4% Average Subordinated AAA Tranche Thickness 13.9% Max: 31.2% Min: 2.4% Number of Transactions that are Amortizing 43 Information used in this presentation is sourced from LoanPerformance except in the following circumstances: • Underlying Reference Obligation data is sourced from Intex • Rating Information is sourced from Bloomberg • Loss Data is sourced from CSFB’s “Locus” system • Sector Categorization is compiled from Individual Transaction Trustee Reports and Moody’s and S&P data “Super Senior” Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Data as of Dec 31
A-11 Summary of Reference Obligations in ourSummary of Reference Obligations in our Transactions:Transactions: Sub Prime Other RMBS CDO CMBS Other ABS Non ABS Total Classifications 63.6% 17.4% 6.4% 7.5% 4.2% 0.9% 100.0% AAA AA A BBB BB <BB NR Rating 2.4% 4.7% 11.6% 55.2% 11.5% 13.6% 1.0% 100.0% Percentages shown are of Gross Transaction Notional “Super Senior” Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary
A-12 Recent Rating PerformanceRecent Rating Performance Underlying Reference Obligations in our Transactions Moody’s S&P % of Sub Prime Upgraded Since Deal Inception 1.2% 0.6% % of Sub Prime Downgraded Since Deal Inception 17.1% 21.0% Overall Transaction Deals with Junior Tranches on Negative Review 27 33 Junior Tranches Downgraded 10 25 AIG Tranche Downgraded None None “Super Senior” Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary
A-13 Underlying Sub Prime Reference ObligationsUnderlying Sub Prime Reference Obligations AAA AA A BBB BB <BB Total Rating 0.5% 1.4% 4.9% 36.4% 9.4% 11.0% 63.6% Pre 2004 2004 2005 2006 2007 Total Vintage 3.3% 17.8% 34.6% 4.4% 3.5% 63.6% Q1 2005 Q2 2005 Q3 2005 Q4 2005 Q1 2006 Q2 2006 Q3 2006 Q4 2006 Q1 2007 Q2 2007 Q3 2007 Q4 2007 Total Vintage by Quarter 13.8% 11.6% 5.9% 3.3% 1.5% 1.3% 0.7% 0.9% 1.4% 1.3% 0.5% 0.3% 42.5% AAA AA A BBB BB <BB Total 2006 Vintage 0.1% 0.4% 0.2% 0.9% 0.6% 2.2% 4.4% AAA AA A BBB BB <BB Total 2007 Vintage 0.2% 0.4% 0.4% 0.7% 0.8% 1.0% 3.5% “Super Senior” Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
A-14 Additional Sub Prime BreakdownAdditional Sub Prime Breakdown Avg. % of Sub Prime Collateral that is 2nd Lien 3.5% Avg. % of Sub Prime Collateral with 2nd Lien >90% 0.8% Average LTV at Inception 80.6 Avg. HPA on Reference Sub Prime Bonds 7.6% Max: 22.8% Min: 3.6% Current Average 12 mos CPR Rate 29.8% Average FICO Score 624 Floating 2 Yr ARM 3 Yr ARM Fixed Average Loan Type 61.3% 5.8% 6.7% 38.7% Sub Prime Originators New Century 5.4% State Concentration California 22.5% Ameriquest 4.8% Florida 7.2% Countrywide 3.7% Texas 6.1% Option 1 3.7% New York 5.8% Fremont 2.9% Michigan 2.0% Current Weighted Average Loss Rate on Sub Prime 1.1% Percentages shown are of Total Sub Prime “Super Senior” Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary
A-15 Underlying CDO Reference Obligations BreakdownUnderlying CDO Reference Obligations Breakdown High Grade Mezz Other ABS CMBS Other Total Classifications 0.7% 3.3% 0.3% 0.8% 1.3% 6.4% AAA AA A BBB BB <BB NR Total Rating 0.2% 0.5% 1.1% 2.5% 0.6% 1.3% 0.2% 6.4% Pre 2003 2003 2004 2005 2006 2007 Total Vintage 0.3% 0.5% 2.0% 2.6% 0.8% 0.2% 6.4% Other RMBS by Vintage Pre 2003 2003 2004 2005 2006 2007 Total 0.2% 2.0% 6.2% 7.2% 1.4% 0.4% 17.4% “Super Senior” Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
A-16 ““Super SeniorSuper Senior”” Multi Sector CDO ExposureMulti Sector CDO Exposure Consisting of Mixed Collateral,Consisting of Mixed Collateral, including Sub Prime:including Sub Prime: High Grade Collateral Underlying SummaryHigh Grade Collateral Underlying Summary
A-17 Summary of Transaction Current StatusSummary of Transaction Current Status Gross Notional 53.1bn AIG Notional Exposure (net of transaction subordination) 43.5bn Max: 2.2 bn Min: 44.8 mm Avg: 964 mm Number of Transactions 45 Managed Transactions 20 Static Transactions 25 Average Number of Obligors 206 Average AIG Attachment Point 15.4% Max: 46.2% Min: 10.0% Average % of AIG Subordination that is AAA Rated 43.0% Max: 86.1% Min: 0.0% Average Subordinated AAA Tranche Thickness 7.4% Max: 39.7% Min: 0.0% Number of Transactions that are Amortizing 32 “Super Senior” Multi Sector CDO Exposure – High Grade Collateral Underlying SummaryData as of Dec 31 Information used in this presentation is sourced from LoanPerformance except in the following circumstances: • Underlying Reference Obligation data is sourced from Intex • Rating Information is sourced from Bloomberg • Loss Data is sourced from CSFB’s “Locus” system • Sector Categorization is compiled from Individual Transaction Trustee Reports and Moody’s and S&P data
A-18 Summary of Underlying Reference Obligations in ourSummary of Underlying Reference Obligations in our Transactions:Transactions: Sub Prime Other RMBS CDO CMBS Other ABS Non ABS Total Classifications 47.5% 27.9% 15.2% 6.8% 2.4% 0.2% 100.0% AAA AA A BBB BB <BB NR Rating 28.2% 35.7% 26.1% 4.1% 2.4% 1.1% 2.4% 100.0% “Super Senior” Multi Sector CDO Exposure – High Grade Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
A-19 Recent Rating PerformanceRecent Rating Performance Underlying Reference Obligations in our Transactions Moody’s S&P % of Sub Prime Upgraded Since Deal Inception 1.1% 0.9% % of Sub Prime Downgraded Since Deal Inception 1.6% 2.3% Overall Transaction Deals with Junior Tranches on Negative Review 9 14 Junior Tranches Downgraded 3 None AIG Tranche Downgraded None None “Super Senior” Multi Sector CDO Exposure – High Grade Collateral Underlying Summary
A-20 Underlying Sub Prime Reference ObligationsUnderlying Sub Prime Reference Obligations AAA AA A BBB BB <BB Total Rating 4.6% 21.4% 18.1% 2.3% 0.8% 0.3% 47.5% Pre 2004 2004 2005 2006 2007 Total Vintage 2.2% 13.7% 26.4% 2.6% 2.6% 47.5% Q1 2005 Q2 2005 Q3 2005 Q4 2005 Q1 2006 Q2 2006 Q3 2006 Q4 2006 Q1 2007 Q2 2007 Q3 2007 Q4 2007 Total Vintage by Quarter 6.6% 9.5% 5.8% 4.5% 1.3% 0.6% 0.3% 0.4% 0.7% 0.8% 0.7% 0.4% 31.6% AAA AA A BBB BB <BB Total 2006 Vintage 0.9% 0.9% 0.5% 0.1% 0.1% 0.1% 2.6% AAA AA A BBB BB <BB Total 2007 Vintage 0.9% 0.9% 0.3% 0.4% 0.1% 0.0% 2.6% Percentages shown are of Gross Transaction Notional “Super Senior” Multi Sector CDO Exposure – High Grade Collateral Underlying Summary
A-21 Additional Sub Prime BreakdownAdditional Sub Prime Breakdown Avg. % of Sub Prime Collateral that is 2nd Lien 3.2% Avg. % of Sub Prime Collateral with 2nd Lien >90% 0.7% Average LTV at Inception 80.6 Avg. HPA on Reference Sub Prime Bonds 7.7% Max: 22.3% Min: 3.1% Current Average 12 month CPR Rate 30.2% Average FICO Score 625 Floating 2 Yr ARM 3 Yr ARM Fixed Average Loan Type 60.2% 5.3% 7.5% 39.8% Sub Prime Originators Countrywide 4.8% State Concentration California 22.9% Ameriquest 4.5% Florida 7.4% New Century 4.0% Texas 5.7% Lehman 2.7% New York 5.7% Fremont 2.7% Michigan 2.0% Current Weighted Average Loss Rate on Sub Prime 1.2% Percentages shown are of Total Sub Prime “Super Senior” Multi Sector CDO Exposure – High Grade Collateral Underlying Summary
A-22 Underlying CDO Reference Obligations BreakdownUnderlying CDO Reference Obligations Breakdown High Grade Mezzanine Other ABS CMBS Other Total Classifications 2.8% 8.0% 0.7% 1.1% 2.6% 15.2% AAA AA A BBB BB <BB NR Total Rating 4.0% 5.1% 2.0% 1.2% 1.4% 0.6% 0.9% 15.2% Pre 2003 2003 2004 2005 2006 2007 Total Vintage 0.8% 1.3% 6.4% 4.8% 1.5% 0.4% 15.2% Other RMBS by Vintage Pre 2003 2003 2004 2005 2006 2007 Total 0.4% 1.0% 7.1% 16.1% 2.3% 1.0% 27.9% Percentages shown are of Gross Transaction Notional “Super Senior” Multi Sector CDO Exposure – High Grade Collateral Underlying Summary
A-23 Cash Multi Sector CDO ExposureCash Multi Sector CDO Exposure Consisting of Mixed Collateral,Consisting of Mixed Collateral, including Sub Prime:including Sub Prime: Mezzanine Collateral Underlying SummaryMezzanine Collateral Underlying Summary
A-24 Summary of Transaction Current StatusSummary of Transaction Current Status AIG Notional Exposure (net of transaction subordination) 2.6bn Max: 191 mm Min: 4 mm Avg: 41 mm Number of Transactions 63 Managed Transactions 21 Static Transactions 42 Average Number of Obligors 162 Average AIG Attachment Point (weighted by transaction notional) 28.5% Max: 59.6% Min: 12.0% Number of Transactions that are Amortizing 34 Rating of Exposure Moody’s S&P AAA 47 58 AA 2 5 NR 14 0 Information used in this presentation is sourced from LoanPerformance except in the following circumstances: • Underlying Collateral Data is sourced from Intex • Rating Information is sourced from Bloomberg • Loss Data is sourced from CSFB’s “Locus” system • Sector Categorization is sourced from Individual Transaction Trustee Reports along with Moody’s and S&P data Cash Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary
A-25 Summary of Reference Obligations in ourSummary of Reference Obligations in our Transactions:Transactions: Sub Prime Other RMBS CDO CMBS Other ABS Non ABS Total Classifications 65.8% 16.7% 4.7% 7.2% 5.0% 0.6% 100.0% AAA AA A BBB BB <BB NR Rating 2.6% 5.7% 13.7% 51.4% 12.4% 13.4% 0.8% 100.0% Cash Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
A-26 Recent Rating PerformanceRecent Rating Performance Underlying Reference Obligations in our Transactions Moody’s S&P % of Sub Prime Upgraded Since Deal Inception 2.2% 1.7% % of Sub Prime Downgraded Since Deal Inception 29.8% 36.2% Overall Transaction Deals with Junior Tranches on Negative Review 18 18 Junior Tranches Downgraded 6 13 AIG Tranche Downgraded None 5 Cash Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary
A-27 Underlying Sub Prime Reference ObligationsUnderlying Sub Prime Reference Obligations AAA AA A BBB BB <BB Total Rating 0.4% 1.2% 6.7% 35.6% 10.5% 11.4% 65.8% Pre 2004 2004 2005 2006 2007 Total Vintage 3.6% 23.1% 32.8% 3.5% 2.8% 65.8% Q1 2005 Q2 2005 Q3 2005 Q4 2005 Q1 2006 Q2 2006 Q3 2006 Q4 2006 Q1 2007 Q2 2007 Q3 2007 Q4 2007 Total Vintage by Quarter 14.3% 10.7% 5.1% 2.7% 1.2% 0.9% 0.5% 0.9% 1.1% 0.9% 0.5% 0.3% 39.1% AAA AA A BBB BB <BB Total 2006 Vintage 0.1% 0.3% 0.1% 0.7% 0.5% 1.8% 3.5% AAA AA A BBB BB <BB Total 2007 Vintage 0.1% 0.3% 0.3% 0.8% 0.6% 0.7% 2.8% Cash Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
A-28 Additional Sub Prime BreakdownAdditional Sub Prime Breakdown Avg. % of Sub Prime Reference Collateral that is 2nd Lien 3.6% Avg. % of Sub Prime Reference Collateral with 2nd Lien >90% 0.9% Average LTV at Inception 80.92 Avg. HPA on Underlying Sub Prime Bonds 8.1% Max: 18.5% Min: 4.0% Current Average 12 mos CPR Rate 30.2% Average FICO Score 626 Floating 2 Yr ARM 3 Yr ARM Fixed Average Loan Type 59.7% 6.3% 6.9% 40.3% Top 5 Sub Prime Originators New Century 6.2% Concentration by State California 22.5% Ameriquest 5.7% Florida 6.8% Countrywide 4.2% Texas 6.3% Fremont 3.4% New York 6.0% Option One 3.4% Michigan 2.1% Current Weighted Average Loss Rate on Sub Prime 1.1% Cash Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Percentages shown are of Total Sub Prime
A-29 Underlying CDO Reference Obligations BreakdownUnderlying CDO Reference Obligations Breakdown High Grade Mezz Other ABS CMBS Other Total Classifications 0.4% 2.4% 0.5% 0.7% 0.7% 4.7% AAA AA A BBB BB <BB NR Total Rating 0.1% 0.5% 0.6% 2.0% 0.4% 1.0% 0.1% 4.7% Pre 2003 2003 2004 2005 2006 2007 Total Vintage 0.2% 0.2% 1.0% 2.4% 0.8% 0.1% 4.7% Other RMBS by Vintage Pre 2003 2003 2004 2005 2006 2007 Total 0.3% 2.5% 7.1% 5.4% 1.2% 0.2% 16.7% Cash Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
A-30 Cash Multi Sector CDO ExposureCash Multi Sector CDO Exposure Consisting of Mixed Collateral,Consisting of Mixed Collateral, including Sub Prime:including Sub Prime: High Grade Collateral Underlying SummaryHigh Grade Collateral Underlying Summary
A-31 Summary of Transaction Current StatusSummary of Transaction Current Status AIG Notional Exposure (net of transaction subordination) 1.5bn Max: 250 mm Min: 14 mm Avg: 134 mm Number of Transactions 11 Managed Transactions 4 Static Transactions 7 Average Number of Obligors 202 Average AIG Attachment Point 15.0% Max: 37.0% Min: 11.4% Number of Transactions that are Amortizing 5 Rating of Exposure Moody’s S&P AAA 11 11 Information used in this presentation is sourced from LoanPerformance except in the following circumstances: • Underlying Collateral Data is sourced from Intex • Rating Information is sourced from Bloomberg • Loss Data is sourced from CSFB’s “Locus” system • Sector Categorization is sourced from Individual Transaction Trustee Reports along with Moody’s and S&P data Cash Multi Sector CDO Exposure – High Grade Collateral Underlying Summary
A-32 Summary of Underlying Reference Obligations in ourSummary of Underlying Reference Obligations in our Transactions:Transactions: Sub Prime Other RMBS CDO CMBS Other ABS Non ABS Total Classifications 39.0% 24.4% 12.6% 20.5% 2.5% 1.0% 100.0% AAA AA A BBB BB <BB NR Rating 37.1 23.5% 22.6% 6.1% 1.9% 1.1% 7.7% 100.0% Cash Multi Sector CDO Exposure – High Grade Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
A-33 Recent Rating PerformanceRecent Rating Performance Underlying Reference Obligations in our Transactions Moody’s S&P % of Sub Prime Upgraded Since Deal Inception 3.8% 6.1% % of Sub Prime Downgraded Since Deal Inception 4.6% 7.4% Overall Transaction Deals with Junior Tranches on Negative Review 2 None Junior Tranches Downgraded None None AIG Tranche Downgraded None None Cash Multi Sector CDO Exposure – High Grade Collateral Underlying Summary
A-34 Underlying Sub Prime Reference ObligationsUnderlying Sub Prime Reference Obligations AAA AA A BBB BB <BB Total Rating 5.3% 15.1% 14.3% 3.4% 0.8% 0.1% 39.0% Pre 2004 2004 2005 2006 2007 Total Vintage 2.8% 9.9% 22.4% 2.9% 1.0% 39.0% Q1 2005 Q2 2005 Q3 2005 Q4 2005 Q1 2006 Q2 2006 Q3 2006 Q4 2006 Q1 2007 Q2 2007 Q3 2007 Q4 2007 Total Vintage by Quarter 3.8% 4.0% 6.2% 8.4% 1.3% 0.5% 0.6% 0.5% 0.2% 0.4% 0.3% 0.1% 26.3% AAA AA A BBB BB <BB Total 2006 Vintage 1.4% 0.6% 0.6% 0.3% 0.0% 0.0% 2.9% AAA AA A BBB BB <BB Total 2007 Vintage 0.3% 0.4% 0.3% 0.0% 0.0% 0.0% 1.0% Cash Multi Sector CDO Exposure – High Grade Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
A-35 Additional Sub Prime BreakdownAdditional Sub Prime Breakdown Avg. % of Reference Transaction Collateral that is 2nd Lien 2.4% Avg. % of Reference Transaction with 2nd Lien >90% 0.4% Average LTV at Inception 80.65 Avg. HPA on Underlying Sub Prime Bonds 6.7% Max: 16.6% Min: 3.1% Current Average 12 month CPR Rate 30.8% Average FICO Score 626 Floating 2 Yr ARM 3 Yr ARM Fixed Average Loan Type 61.5% 4.4% 7.9% 38.5% Top 5 Sub Prime Originators Countrywide 8.3% State Concentration California 24.3% Ameriquest 5.0% Florida 7.7% Lehman 5.1% Texas 5.4% New Century 4.3% New York 5.4% First Franklin 3.3% Michigan 1.7% Current Weighted Average Loss Rate on Sub Prime 1.0% Cash Multi Sector CDO Exposure – High Grade Collateral Underlying Summary Percentages shown are of Total Sub Prime
A-36 Underlying CDO Reference Obligations BreakdownUnderlying CDO Reference Obligations Breakdown High Grade Mezzanine Other ABS CMBS Other Total Classifications 2.0% 4.8% 1.1% 2.2% 2.5% 12.6% AAA AA A BBB BB <BB NR Total Rating 5.0% 2.2% 1.8% 1.3% 0.9% 0.5% 0.9% 12.6% Pre 2003 2003 2004 2005 2006 2007 Total Vintage 1.0% 2.4% 4.0% 2.5% 2.0% 0.7% 12.6% Other RMBS by Vintage Pre 2003 2003 2004 2005 2006 2007 Total 2.2% 2.1% 6.1% 11.1% 2.2% 0.7% 24.4% Cash Multi Sector CDO Exposure – High Grade Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
AIG Conference Call Credit Presentation - Supplemental Materials - February 29, 2008

More Related Content

PDF
Steering with Expected Loss
László Árvai
 
PDF
Steering with Expected Loss
László Árvai
 
PDF
IFRS 9 : Accounting Meets Risk Management by En Shah Zain
Albakry Azis
 
PDF
IFRS 9 conference presentation - Philip Lewis
Aptitude Software
 
PPT
Methods of Bank Performance Evaluation
Neha Agrawal
 
PPT
Accounting std1
Nirbhay Singh
 
PPTX
Accounting std1
Nirbhay Singh
 
PDF
IFRS9 white paper - Credit Today and HML
HML Ltd
 
Steering with Expected Loss
László Árvai
 
Steering with Expected Loss
László Árvai
 
IFRS 9 : Accounting Meets Risk Management by En Shah Zain
Albakry Azis
 
IFRS 9 conference presentation - Philip Lewis
Aptitude Software
 
Methods of Bank Performance Evaluation
Neha Agrawal
 
Accounting std1
Nirbhay Singh
 
Accounting std1
Nirbhay Singh
 
IFRS9 white paper - Credit Today and HML
HML Ltd
 

What's hot (18)

PDF
9 challenges of the IFRS 9 - Impairments
Peter Plochan, FRM
 
PPT
Ch 13
kpserver
 
PPTX
Camels approach
Vishal Parmar
 
PPTX
Credit Impairment under IFRS 9 for Banks
Faraz Zuberi
 
PDF
Rma May22 Stress Testing In The Context Of Icaap
Guan Khoo
 
PDF
IFRS 9 Model Risk Management - Given the Short Shift ?
Sandip Mukherjee CFA, FRM
 
PPT
FINANCIAL MANAGEMENT PPT BY FINMANFin man report mark lester samonte
Mary Rose Habagat
 
PPT
Operational Risk &amp; Basel Ii
jhsiddiqi2003
 
PPTX
Investment decisions
PPTMBA1
 
PPTX
Securities Financing Transactions Regulation
Megha Gupta
 
PPT
Capital Budgeting
Dayasagar S
 
PPTX
Fm ak
AKBARKHAN PATHAN
 
PPTX
Ak finaacial
AKBARKHAN PATHAN
 
PDF
Operational Risk Management under BASEL era
Treat Risk
 
PPTX
Risk and return
PPTMBA1
 
PPTX
Camels Rating
Parveen Bari
 
PPTX
Camel Analysis of IDCs Rank of Financial Ratios
John Rickmeier
 
DOCX
Capital rationing
Shifas ibrahim MBA student @ ILAHIA SCHOOL OF MANAGEMENT STUDIES
 
9 challenges of the IFRS 9 - Impairments
Peter Plochan, FRM
 
Ch 13
kpserver
 
Camels approach
Vishal Parmar
 
Credit Impairment under IFRS 9 for Banks
Faraz Zuberi
 
Rma May22 Stress Testing In The Context Of Icaap
Guan Khoo
 
IFRS 9 Model Risk Management - Given the Short Shift ?
Sandip Mukherjee CFA, FRM
 
FINANCIAL MANAGEMENT PPT BY FINMANFin man report mark lester samonte
Mary Rose Habagat
 
Operational Risk &amp; Basel Ii
jhsiddiqi2003
 
Investment decisions
PPTMBA1
 
Securities Financing Transactions Regulation
Megha Gupta
 
Capital Budgeting
Dayasagar S
 
Fm ak
AKBARKHAN PATHAN
 
Ak finaacial
AKBARKHAN PATHAN
 
Operational Risk Management under BASEL era
Treat Risk
 
Risk and return
PPTMBA1
 
Camels Rating
Parveen Bari
 
Camel Analysis of IDCs Rank of Financial Ratios
John Rickmeier
 
Capital rationing
Shifas ibrahim MBA student @ ILAHIA SCHOOL OF MANAGEMENT STUDIES
 
Ad

Viewers also liked (20)

PDF
morgan stanley Earnings Archive2003 2nd
finance2
 
PDF
JPMorgan Chase 2008 Proxy Statement
finance2
 
DOC
Shantanu Dhanuka
Shantanu Dhanuka
 
PPTX
Presentation 3
orlando alonzo
 
PDF
Архитектор информационных систем
Elena7436
 
PDF
RMG Design Build Protect
Mercedes Keene
 
PPTX
Winters mindfulness
jaimiallison
 
PPT
Ética Profesional
Atef Alín Akram Adel
 
PDF
Jul 15_HungryTraveller
Clare Winskill
 
PDF
Hình vẽ đơn giản giúp nhà trở nên vui mắt
Thi công sơn giá rẻ
 
PPTX
Marko Nikolic - impress your customers - easy or not
Marko Nikolic
 
PPSX
Ict training
Bình Nguyễn Duy
 
PDF
IBM Historical IBM quarterly dividend rates and payable and record dates.
finance2
 
PDF
2003 JP Morgan Healthcare Conference
finance2
 
PDF
Verizon Reports Sustained Revenue Growth and Continued Strong Cash Flows fo...
finance2
 
PDF
morgan stanley 10-K Disclosure
finance2
 
PDF
Aust dp
koutly
 
PDF
VACO Raleigh Team
jchriswalker
 
PPTX
Pop up bibliotheek
Koninklijke Bibliotheek
 
PPT
Kelley's slides
cookie6
 
morgan stanley Earnings Archive2003 2nd
finance2
 
JPMorgan Chase 2008 Proxy Statement
finance2
 
Shantanu Dhanuka
Shantanu Dhanuka
 
Presentation 3
orlando alonzo
 
Архитектор информационных систем
Elena7436
 
RMG Design Build Protect
Mercedes Keene
 
Winters mindfulness
jaimiallison
 
Ética Profesional
Atef Alín Akram Adel
 
Jul 15_HungryTraveller
Clare Winskill
 
Hình vẽ đơn giản giúp nhà trở nên vui mắt
Thi công sơn giá rẻ
 
Marko Nikolic - impress your customers - easy or not
Marko Nikolic
 
Ict training
Bình Nguyễn Duy
 
IBM Historical IBM quarterly dividend rates and payable and record dates.
finance2
 
2003 JP Morgan Healthcare Conference
finance2
 
Verizon Reports Sustained Revenue Growth and Continued Strong Cash Flows fo...
finance2
 
morgan stanley 10-K Disclosure
finance2
 
Aust dp
koutly
 
VACO Raleigh Team
jchriswalker
 
Pop up bibliotheek
Koninklijke Bibliotheek
 
Kelley's slides
cookie6
 
Ad

Similar to AIG Conference Call Credit Presentation - Supplemental Materials - February 29, 2008 (20)

PDF
PwC CECL Overview Placemat_Final
Ben Havird
 
PPTX
Financial and technical regulations issued by insurance authority
Aftab Hasan
 
PPTX
CECL becomes effective in 2020
Allen, Gibbs & Houlik, L.C.
 
PPTX
NFRS vs Nepal Rastra Bank Directives
Dev Mukunda
 
PDF
Outsourcing GIA Accounting whitepaper 2016
Rich Lawrence
 
PDF
Ratio analysiss
Jason Spring
 
PDF
WSO2CON 2024 - Lessons from the Field: Legacy Platforms – It's Time to Let Go...
WSO2
 
PDF
Impairment Presentation 2015
Grahame Nanson
 
PDF
fiinratings-corporate-credit-rating-methodology.pdf
ssuser8909b62
 
PDF
9 dan 10 Intepreting Financial Statement and Marketing Decision.pdf
BambangPamungkas32
 
PPT
200922440738781.ppt
vineetranjan22
 
PPT
caiib_rmmodd_nov08.ppt
Pranith roy
 
PDF
The Age of Alignment Part III: Moving From Theory to Practice
Pearl Meyer
 
PDF
Overview of Corporate Finance in India a presentation
footydigarse
 
PPT
risk.ppt
mohamedmustafa854838
 
PPTX
Capital Budgeting PPT.pptx
adnankhan765563
 
PPT
Risk Management & Banks Analytics & Information Requirement
Tuhin Choudhury
 
PPTX
International Financial Statement Analysis.pptx
MohamedAbdi347025
 
PDF
GEO NECF 2015 - Best Practices and Trends in Financial Reporting
Andrea Huck-Esposito
 
PPT
02_lecture 8.ppt
adnankhan765563
 
PwC CECL Overview Placemat_Final
Ben Havird
 
Financial and technical regulations issued by insurance authority
Aftab Hasan
 
CECL becomes effective in 2020
Allen, Gibbs & Houlik, L.C.
 
NFRS vs Nepal Rastra Bank Directives
Dev Mukunda
 
Outsourcing GIA Accounting whitepaper 2016
Rich Lawrence
 
Ratio analysiss
Jason Spring
 
WSO2CON 2024 - Lessons from the Field: Legacy Platforms – It's Time to Let Go...
WSO2
 
Impairment Presentation 2015
Grahame Nanson
 
fiinratings-corporate-credit-rating-methodology.pdf
ssuser8909b62
 
9 dan 10 Intepreting Financial Statement and Marketing Decision.pdf
BambangPamungkas32
 
200922440738781.ppt
vineetranjan22
 
caiib_rmmodd_nov08.ppt
Pranith roy
 
The Age of Alignment Part III: Moving From Theory to Practice
Pearl Meyer
 
Overview of Corporate Finance in India a presentation
footydigarse
 
risk.ppt
mohamedmustafa854838
 
Capital Budgeting PPT.pptx
adnankhan765563
 
Risk Management & Banks Analytics & Information Requirement
Tuhin Choudhury
 
International Financial Statement Analysis.pptx
MohamedAbdi347025
 
GEO NECF 2015 - Best Practices and Trends in Financial Reporting
Andrea Huck-Esposito
 
02_lecture 8.ppt
adnankhan765563
 

More from finance2 (20)

PDF
The Home Depot Celebrates Hispanic Culture Through Color and Paint With Color...
finance2
 
PDF
The Home Depot and AARP Launch Nationwide Workshops
finance2
 
PDF
Nearly Half of Americans Fail to Check Home Safety Devices at Daylight-Saving...
finance2
 
PDF
View Summary The Home Depot Celebrates the Olympic Spirit With Special Kids...
finance2
 
PDF
The Home Depot Announces First Quarter Results
finance2
 
PDF
home depot Craig Menear Presentation
finance2
 
PDF
home depot 2008 Annual Meeting of Stockholders
finance2
 
PDF
home depot Transcript
finance2
 
PDF
home depot PDF Carol Tomé Presentation
finance2
 
PDF
home depot Paul Raines Presentation
finance2
 
PDF
home depot http://ir.homedepot.com/common/download/download.cfm?companyid=HD&...
finance2
 
PDF
home depot Annual Report 1982
finance2
 
PDF
home depot PDF Frank Blake Presentation
finance2
 
PDF
home depotForward Looking & Non-GAAP Disclosures
finance2
 
PDF
home depot Bank of America 38th Annual Investment Conference
finance2
 
PDF
2008 Home Depot Inc. Earnings Conference
finance2
 
PDF
home depot Quarterly Earnings 2008 3rd
finance2
 
PDF
home depot Annual Report 1983
finance2
 
PDF
home depot Annual Report 1984
finance2
 
PDF
home depot Annual Report 1985
finance2
 
The Home Depot Celebrates Hispanic Culture Through Color and Paint With Color...
finance2
 
The Home Depot and AARP Launch Nationwide Workshops
finance2
 
Nearly Half of Americans Fail to Check Home Safety Devices at Daylight-Saving...
finance2
 
View Summary The Home Depot Celebrates the Olympic Spirit With Special Kids...
finance2
 
The Home Depot Announces First Quarter Results
finance2
 
home depot Craig Menear Presentation
finance2
 
home depot 2008 Annual Meeting of Stockholders
finance2
 
home depot Transcript
finance2
 
home depot PDF Carol Tomé Presentation
finance2
 
home depot Paul Raines Presentation
finance2
 
home depot http://ir.homedepot.com/common/download/download.cfm?companyid=HD&...
finance2
 
home depot Annual Report 1982
finance2
 
home depot PDF Frank Blake Presentation
finance2
 
home depotForward Looking & Non-GAAP Disclosures
finance2
 
home depot Bank of America 38th Annual Investment Conference
finance2
 
2008 Home Depot Inc. Earnings Conference
finance2
 
home depot Quarterly Earnings 2008 3rd
finance2
 
home depot Annual Report 1983
finance2
 
home depot Annual Report 1984
finance2
 
home depot Annual Report 1985
finance2
 

Recently uploaded (20)

PDF
NewBase 12 August 2025 Energy News issue - 1812 by Khaled Al Awadi_compresse...
Khaled Al Awadi
 
PDF
Module 3 - Functions of the Supervisor - Part 1 - Student Resource (1).pdf
OmoobaOrun1
 
PDF
Comments on Crystal Cloud and Energy Star.pdf
Brij Consulting, LLC
 
PDF
Power and position in leadershipDOC-20250808-WA0011..pdf
meghavinikumar2006
 
PDF
TyAnn Osborn: A Visionary Leader Shaping Corporate Workforce Dynamics
CIO WOMEN MAGAIZNE CIO WOMEN MAGAIZNE
 
PDF
Outsourced Audit & Assurance in USA Why Globus Finanza is Your Trusted Choice
Globus finanza
 
PDF
How to Get Business Funding for Small Business Fast
Jake Thornhill
 
PPT
340036916-American-Literature-Literary-Period-Overview.ppt
carinaherdiles611
 
DOCX
Business Management - unit 1 and 2
anithak410
 
PPTX
Board-Reporting-Package-by-Umbrex-5-23-23.pptx
Pars Six Sigma Excellence
 
PPTX
Principles of Marketing, Industrial, Consumers,
efranciscaantoinette
 
PDF
Digital Marketing & E-commerce Certificate Glossary.pdf.................
RoobaV2
 
PDF
Charisse Litchman: A Maverick Making Neurological Care More Accessible
The lifescience magaizne
 
PPT
Chapter four Project-Preparation material
argachewbochena1
 
PPTX
DMT - Profile Brief About Business .pptx
AkhileshKumar724847
 
PDF
Hindu Circuler Economy - Model (Concept)
Avijit Kumar Roy
 
PPTX
Probability Distribution, binomial distribution, poisson distribution
gayusakktivel
 
PPTX
Business Ethics - An introduction and its overview.pptx
Keerthana Chinnathambi
 
PPTX
2025 Product Deck V1.0.pptxCATALOGTCLCIA
AndreaRominaHoyosSur2
 
PDF
pdfcoffee.com-opt-b1plus-sb-answers.pdfvi
thaoonguyenn2311
 
NewBase 12 August 2025 Energy News issue - 1812 by Khaled Al Awadi_compresse...
Khaled Al Awadi
 
Module 3 - Functions of the Supervisor - Part 1 - Student Resource (1).pdf
OmoobaOrun1
 
Comments on Crystal Cloud and Energy Star.pdf
Brij Consulting, LLC
 
Power and position in leadershipDOC-20250808-WA0011..pdf
meghavinikumar2006
 
TyAnn Osborn: A Visionary Leader Shaping Corporate Workforce Dynamics
CIO WOMEN MAGAIZNE CIO WOMEN MAGAIZNE
 
Outsourced Audit & Assurance in USA Why Globus Finanza is Your Trusted Choice
Globus finanza
 
How to Get Business Funding for Small Business Fast
Jake Thornhill
 
340036916-American-Literature-Literary-Period-Overview.ppt
carinaherdiles611
 
Business Management - unit 1 and 2
anithak410
 
Board-Reporting-Package-by-Umbrex-5-23-23.pptx
Pars Six Sigma Excellence
 
Principles of Marketing, Industrial, Consumers,
efranciscaantoinette
 
Digital Marketing & E-commerce Certificate Glossary.pdf.................
RoobaV2
 
Charisse Litchman: A Maverick Making Neurological Care More Accessible
The lifescience magaizne
 
Chapter four Project-Preparation material
argachewbochena1
 
DMT - Profile Brief About Business .pptx
AkhileshKumar724847
 
Hindu Circuler Economy - Model (Concept)
Avijit Kumar Roy
 
Probability Distribution, binomial distribution, poisson distribution
gayusakktivel
 
Business Ethics - An introduction and its overview.pptx
Keerthana Chinnathambi
 
2025 Product Deck V1.0.pptxCATALOGTCLCIA
AndreaRominaHoyosSur2
 
pdfcoffee.com-opt-b1plus-sb-answers.pdfvi
thaoonguyenn2311
 

AIG Conference Call Credit Presentation - Supplemental Materials - February 29, 2008

  • 1. Conference Call Credit Presentation Supplemental Materials Financial Results for the Year Ended December 31, 2007 February 29, 2008 (Revised as to slide 30 and Capital Markets - Data Appendices)
  • 2. 1 It should be noted that this presentation and the remarks made by AIG representatives may contain projections concerning financial information and statements concerning future economic performance and events, plans and objectives relating to management, operations, products and services, and assumptions underlying these projections and statements. It is possible that AIG's actual results and financial condition may differ, possibly materially, from the anticipated results and financial condition indicated in these projections and statements. Factors that could cause AIG's actual results to differ, possibly materially, from those in the specific projections and statements are discussed in Item 1A. Risk Factors of AIG's Annual Report on Form 10-K for the year ended December 31, 2007. AIG is not under any obligation (and expressly disclaims any such obligations) to update or alter its projections and other statements whether as a result of new information, future events or otherwise. This presentation may also contain certain non-GAAP financial measures. The reconciliation of such measures to the comparable GAAP figures are included in the Fourth Quarter 2007 Financial Supplement available in the Investor Information Section of AIG's corporate website, www.aigcorporate.com.
  • 4. 3 Outline • Business Rationale, Portfolio Composition & Underwriting Standards • Fundamental Risk Assessment & Stress Testing • Accounting, Valuation Fundamentals & Economic Capital • Conclusions & Next Steps
  • 5. 4 Outline • Business Rationale, Portfolio Composition & Underwriting Standards • Fundamental Risk Assessment & Stress Testing • Accounting, Valuation Fundamentals & Economic Capital • Conclusions & Next Steps
  • 6. 5 • No uniform definition for “Super Senior” risk across the market. • AIGFP defines “Super Senior” risk as a risk where there is no expected loss at contract inception, even under its conservative stress assumptions. • Due diligence and AIGFP proprietary modeling incorporates significantly more conservative assumptions, including for recovery rates, than those used by the rating agencies. • While rating agency models and attachment points are useful verification tools, AIGFP always builds and models each “Super Senior” transaction with its own more conservative assumptions. • The attachment point for the “Super Senior” portion of each portfolio is modeled as a minimum threshold above which there is no expected loss to AIGFP. The final attachment point is negotiated to exceed the modeled attachment point, giving AIGFP an additional cushion of subordination to its risk position. AIG Financial Products “Super Senior” Credit Default Swap (CDS) Business
  • 7. 6 Objectives for Counterparties in “Super Senior” CDS Transactions • Regulatory capital relief • Facilitate securitization • Arbitrage
  • 8. 7 Summary Statistics “Super Senior” Credit Derivatives* Transaction Type Corporate – Regulatory Capital Motivated European Residential Mortgage – Regulatory Capital Motivated Corporate – Arbitrage Motivated Multi-Sector CDOs Transactions w/Mixed Collateral including Subprime Transactions w/No Subprime Total Multi- Sector CDOs Gross Notional ($ Billion) 306.0 182.8 87.3 82.8 27.3 110.1 AIGFP Net Notional Exposure ($ Billion) 229.6 149.1 70.4 61.4 16.8 78.2 Number of Transactions 58 35 36 103 13 116 Weighted Average Subordination (%) ¹ 22.0% 13.8% 18.3% 23.3% 18.0% 21.9% Weighted Average Number of Obligors / Transaction 1,571 74,819 122 194 185 192 Expected Maturity (Years) 1.2 ² 2.3 ² 4.0 5.0 ³ 5.4 ³ 5.1 ³ 1. Weighted by Gross Transaction Notional 2. Maturity shown reflects first non-regulatory call date, although majority of transactions have Regulatory Capital Calls from Jan 08 3. Reflects the Weighted Average Life*All data is as of December 31, 2007.
  • 9. 8 Credit Underwriting – Regulatory Capital Corporate Transactions Every Transaction Passes Through the Same Process of Careful Selection and Due Diligence • For each regulatory capital-motivated transaction, we review the originating bank’s underwriting standards, including for example: – The bank’s internal rating procedures, its construction and criteria, along with its application to both loan and obligor ratings; – The extent to which account officers are empowered to make lending decisions and/or overrule any scoring system; – All loss mitigation and foreclosure strategies; – The bank’s internal rating system to ensure it is seasoned enough to enable it to have built transition matrices to help validate the ratings; and – The mapping of the bank’s internal rating scales to those of the rating agencies and confirmation of the mapping through discussion with the bank and the rating agencies. • Investigation of any industry or geographic concentrations in the loan pool and review of rationale and any potentially mitigating factors. We strive to create diverse and granular pools of credits. • Where possible, we undertake an analysis of each name in the portfolio to assign each an internal rating (‘AIGFP rating’). The AIGFP rating (which in all cases is equal to or lower than those publicly assigned) is used for modeling purposes. • Price negotiation is bespoke and reflects the counterparty’s unique position and capital requirements. With the implementation of Basel II, such pricing typically reflects an agreed minimum fee due to the very short expected life of the transaction.
  • 10. 9 Credit Underwriting – Regulatory Capital European Residential Mortgage Transactions • Understanding of the bank’s lending philosophy and the extent (if any) to which this has changed in recent times • Underwriting standards and credit approval procedures, including: – How many and which individual officers are empowered to make lending decisions – Loan size that can be approved by different individuals/groups – The extent to which the approval system has been automated and, if so, the procedures for permitting any overrides – What criteria do the credit officers or automated system use to make the lending decision • Loss and recovery experience of the bank • Historical experience - annual losses over the past 10 years or as far back as the bank’s records allow • Monitoring of delinquencies, work-out and recovery procedures • Groups and procedures in place to monitor and manage delinquencies • Bank policies on work-out and foreclosure • Price negotiation is bespoke, similar to regulatory capital corporate transactions. Thorough Review of all Underwriting, Processes and Performance
  • 11. 10 Credit Underwriting – Arbitrage Motivated Corporate Transactions Every Transaction Passes Through the Same Process of Careful Selection and Due Diligence • Similar to the Regulatory Capital transactions, an internal AIGFP rating is assigned to each name in the portfolio (which in all cases is equal to or lower than those publicly assigned) and used for modeling purposes. • Removal of any stressed credits or those which, in our view, have particularly unfavorable outlooks. • Review of current market spreads for each name, where available, to ensure rating used reflects all currently available information. • Selection of credits so as to ensure a diverse and granular pool both in terms of industry and geography. • Pricing discussion based around both the current spread environment and the counterparty’s specific hedging requirements at that particular time.
  • 12. 11 Past Performance Corporate Loans and European Residential Mortgages AIGFP Net Notional Exposure ($ Billions) Total Losses in Reference Pool to Date Weighted Average Attachment Point Multiple of Losses Required Before AIGFP Has any Loss1 Corporate Loans – Regulatory Capital $229.6 0.01% 22.0% 2,234x European Residential Mortgages – Regulatory Capital $149.1 0.03% 13.8% 487x Corporate Loans – Arbitrage $70.4 0.22% 18.3% 81x • The losses realized to date in the underlying reference pools have been extremely small relative to our attachment points. This is a reflection of the positive selection of the portfolios, the parties’ motivations and the highly conservative modeling. AIGFP has not incurred any realized losses from the underlying collateral in these pools. December 31, 2007 Extremely Low Loss Rates in Reference Pools with Attachment Points Always Significantly Higher 1.1. Represents multiple of total losses in reference pool to date reRepresents multiple of total losses in reference pool to date required to exhaust the entire subordination structure before AIGFquired to exhaust the entire subordination structure before AIGFP incurs a loss.P incurs a loss.
  • 13. 12 Any realized credit losses are allocated sequentially: Equity, BB, BBB, A, AAA, then “Super Senior” Underlying portfolio typically comprises 125-200 obligations from various sectors. Those obligations typically have their own subordination embedded “Super Senior” Risk Layer - AIGFP Net Notional Exposure Typical Tranche Structure of a Multi-Sector CDO Including “Super Senior” Layer Equity BB BBB A AAA AAA A AA BBB AAA A AA BBB AAA A AA BBB AAA A AA BBB AAA A AA BBB AAA A AA BBB AAA AA A BBB AAA AA A BBB AAA AA A BBB AAA AA A BBB AAA AA A BBB Portfolio tranched into different risk layers AIGFP Attachment Point Gross Transaction Notional Residential and commercial mortgages, auto loans, etc., are securitized Specific individually rated tranches from those securitizations are purchased by the CDO The CDO is tranched into different layers of risk with the “Super Senior” layer being the most risk remote Protection buyer makes periodic payments to protection seller who in turn makes payments if losses, which are allocated sequentially, exceed the relevant subordination
  • 14. 13 • Review and analysis of the CDO Manager (on all transactions), including: – Review of personnel and their experience and suitability for managing the assets and the structure – Track record and past performance of the manager in all asset classes – Risk retention / incentive policy in place for key employees • Review and analysis of the entire collateral portfolio, including: – The eligibility criteria for all securities – The proposed single security / obligor concentration limits – Geographic portfolio diversification – Sector / industry portfolio diversification – Maturity / expected amortization profile of the assets and the portfolio – Review of agency ratings of securities and portfolio weighted average rating factor – Currency and interest rate exposures and hedging requirements Credit Underwriting – Multi-Sector CDO Transactions Highly Selective Review of Manager, Collateral and Proposed Structure
  • 15. 14 • Analysis of the key transaction terms, including: – The term of any proposed re-investment period – Management trading discretions, if any – Portfolio quality triggers in place – Over-Collateralization (O/C) and Interest Coverage (I/C) tests – Early amortization events and required procedures Credit Underwriting – Multi-Sector CDO Transactions Highly Selective Review of Manager, Collateral and Proposed Structure
  • 16. 15 Multi-Sector CDOs What Differentiates our Transactions? • Strong adherence to proven conservative underwriting approach. • Highly diversified and granular pools of positively selected reference obligations. • Attachment points are calculated to be extremely remote, frequently with significant AAA-rated tranches below our position. • Calculated attachment points are only a minimum and are non-negotiable. • Extensive due diligence carried out before any transaction is agreed. • Conservative portfolio and obligor limits. • Conservative modeling – Conservative assumptions used for portfolio construction – Significant hair-cuts and stresses applied to inputs Careful Portfolio Selection Combined with Remote Attachment Points
  • 17. 16 Outline • Business Rationale, Portfolio Composition & Underwriting Standards • Fundamental Risk Assessment & Stress Testing • Accounting, Valuation Fundamentals & Economic Capital • Conclusions & Next Steps
  • 18. 17 Fundamental Risk Assessment: AIGFP Credit Review Process • AIGFP’s highly experienced credit team as part of its role conducts continual surveillance around the Super Senior portfolio. • Each quarter AIGFP re-runs its stress models against the entire Super Senior portfolio to account for updated subordination information, ratings migration, delinquencies, defaults and losses. They update, evaluate and stress results relative to the current subordinated layers to assess potential credit quality migration. • The AIGFP global credit team meets quarterly to review the portfolio in depth and the results of the stress model. • They review any deals that show early signs of stress and evaluate the factors leading to any portfolio deterioration to determine whether exposure hedging should be recommended or other actions should be instituted, such as meetings with collateral managers or bank lenders. • The credit review also looks at rating agency changes, early deal terminations and credit trends.
  • 19. 18 Fundamental Risk Assessment AIG ERM • Every quarter AIG ERM reviews AIGFP’s “Super Senior” credit derivative exposures. • The review considers delinquency, defaults and realized loss trends for each transaction relative to updated subordination levels. The assessment includes a review of rating agency actions. It also considers adverse economic and sector trends, where applicable. • ERM identifies all transactions that show any unexpected deterioration or heightened risk and adds them to the internal AIG Watch List. • ERM has initiated a regular process to run stress tests of the multi-sector CDO portfolio to determine if any transactions could pose a risk of realizing a loss if economic conditions deteriorate beyond expectations. • ERM also assesses whether any transactions could represent probable loss, thus potentially requiring the establishment of credit reserves (none to date).
  • 20. 19 Rating Agency Actions • All three major rating agencies continue to rate 89.6% of AIGFP’s $61.4 billion super senior credit derivative multi-sector CDO portfolio with sub-prime RMBS collateral at AAA levels. This is despite massive numbers of CDO downgrades during 2007 and early 2008. • Through February 26, 2008 and based on 12/31/07 multi-sector CDO exposure data, approximately $6.4 billion (10.5%) of the portfolio had been downgraded, mostly by only one agency, as follows (using the lowest rating of any of the three agencies): – $2.3 billion to AA grades; – $2.0 billion to single A grades; and – $2.1 billion to BBB grades • At least one rating agency continues to rate all $6.4 billion downgraded CDO SS tranches as AAA. • Another $8.7 billion of AAA rated super senior CDOs (twelve transactions) currently are on credit watch, mostly by S&P. • No exposure is rated below BBB-.
  • 21. 20 Stress Testing - Illustration of Potential Losses on AIGFP’s “Super Senior" Credit Derivative Portfolio on Multi-Sector CDOs 0.90 11.25 0.00 2.00 4.00 6.00 8.00 10.00 12.00 Modeled Severe Stress Scenario Realizable Loss Unrealized Market Valuation Loss Carried on GAAP Balance Sheet The December 31, 2007 unrealized market valuation loss of $11.25 billion significantly exceeds even a severe modeled realizable portfolio loss. Value of Pre Tax Loss Estimates* $ BN Description of ERM Severe Stress Scenario* Collateral Securities Severe Stress Scenario Q1-Q4 ’07 Subprime RMBS 100% of AA+ or lower Q3-Q4 ’06 Subprime RMBS 100% of AA+ or lower Q1-Q2 ’06 Subprime RMBS 50% of AA+, AA, AA-; 100% of A+ or lower Q3-Q4 ’05 Subprime RMBS 50% of BBB+ or lower Q1-Q2 ’05 Subprime RMBS 100% of BB+ or lower Inner CDOs of ABS 100% of A+ or lower CY’06 & CY’07 Alt-A 100% of A+ or lower *As of December 31, 2007. These stresses are “static” stresses, assumed to result in immediate portfolio loss and do not take any benefit for cash flow diversion and other mitigants.
  • 22. 21 Market Valuation vs. Stress Testing Illustrations • Portfolio Level – Slide 21 shows that market valuations indicate losses that far exceed severe stress scenario realizable losses at total portfolio level. • Transaction Level – Slides 22 to 24 show loss comparisons at transaction level: In all CasesIn all Cases Market valuations indicate losses that exceedMarket valuations indicate losses that exceed severe stress scenario realizable losses forsevere stress scenario realizable losses for each of the transactionseach of the transactions Market valuations indicate losses while severeMarket valuations indicate losses while severe stress scenario realizable losses, generally dostress scenario realizable losses, generally do not even breach the subordination layersnot even breach the subordination layers -- ILLUSTRATIONILLUSTRATION -- -- ILLUSTRATIONILLUSTRATION -- 0.0 1.0 2.0 $Billions In ManyIn Many CasesCases0.0 1.0 2.0 $Billions In Most CasesIn Most Cases AIGFP Net Notional Exposure Subordination Layer Severe Stress Scenario Realizable – Gross Notional Loss Unrealized Market Valuation Loss
  • 23. 22 Stress Testing - Illustration of Potential Losses on AIGFP’s “Super Senior” Credit Derivative Portfolio on Multi-Sector CDOs $ Billions Total AIGFP Net Notional Exposure: $78.2 BN Unrealized Market Valuation Loss: $11.25 BN Severe Stress Realizable Loss: $0.90 BN Attachment Point Subordination Layer “Super Senior” Total Multi-Sector CDO Portfolios Unrealized Market Valuation Loss Severe Stress Scenario Realizable Loss
  • 24. 23 Multi-Sector CDO Portfolio Transactions 0.0 1.0 2.0 3.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 $Billions AIGFP Net Notional Exposure Subordination Layer Severe Stress Scenario Realizable - Gross Notional Loss Unrealized Market Valuation Loss Transaction by Transaction Illustration
  • 25. 24 Multi-Sector CDO Portfolio Transactions 0.0 1.0 2.0 3.0 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 $Billions AIGFP Net Notional Exposure Subordination Layer Severe Stress Scenario Realizable - Gross Notional Loss Unrealized Market Valuation Loss Transaction by Transaction Illustration
  • 26. 25 Multi-Sector CDO Portfolio Transactions 0.0 1.0 2.0 3.0 4.0 5.0 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 $Billions AIGFP Net Notional Exposure Subordination Layer Severe Stress Scenario Realizable - Gross Notional Loss Unrealized Market Valuation Loss Transaction by Transaction Illustration
  • 27. 26 Current Market Implied Probabilities of Default for AIGFP’s “Super Senior” Credit Derivative Portfolio on Multi-Sector CDOs 53% 26% 2%0% 10% 20% 30% 40% 50% 60% Estimate Derived From Historical Default Probabilities Current Exit Value Implied By Market Prices Severe Stress Scenario Implied Probabilities of Default of Gross Exposures Over Lifetime of Portfolio* Dec 2006 Dec 2007 Dec 2007 Market Implied Probabilities of Default Reflect: • Basic Credit Risk Fundamentals; plus • Extreme Liquidity Premium; • Market-Driven Risk Aversion; etc. Current market implied probabilities of default, being orders of magnitude greater than historical probabilities of default, suggest the existence of significant factors included in market prices in addition to credit risk, the principal risk to which AIGFP is exposed. * Implies a weighted average recovery rate of 40% across all por* Implies a weighted average recovery rate of 40% across all portfolios.tfolios.
  • 28. 27 Conclusions From Fundamental Risk Assessment & Stress Testing • All three major rating agencies continue to rate 89.6% of AIGFP’s $61.4 billion super senior CDS multi-sector CDO portfolio with sub-prime RMBS collateral at AAA levels. This is despite massive numbers of CDO downgrades during 2007 and early 2008. No exposure is rated below BBB-. • The December 31, 2007 unrealized market valuation loss of $11.25 billion significantly exceeds even a severe modeled realizable portfolio loss. • Market valuations indicate losses that exceed severe stress scenario losses for each of the transactions. For most of the transactions, market valuations indicate losses, while severe stress scenario realizable losses generally do not even breach the subordination layers. • Current market implied probabilities of default, being orders of magnitude greater than historical probabilities of default, suggest the existence of significant factors included in the market prices in addition to credit risk, the principal risk to which AIGFP is exposed. • AIGFP wrote credit derivative protection as a principal based upon sound underwriting procedures and has the ability and intent to hold its positions until contract maturity or call by the counterparty.
  • 29. 28 Outline • Business Rationale, Portfolio Composition & Underwriting Standards • Fundamental Risk Assessment & Stress Testing • Accounting, Valuation Fundamentals & Economic Capital • Conclusions & Next Steps
  • 30. 29 Accounting for “Super Senior” Credit Derivative Swaps • AIGFP accounts for its “Super Senior” Credit Derivative portfolio in accordance with FAS 133 and EITF 02-3: – At inception the credit derivative is recorded at its transaction price as that is the best indicator of fair value. – Subsequent changes in fair value are recognized in earnings. • Through June 30, 2007 there was minimal change in fair value since the inception of the derivatives: – The “Super Senior” credit derivative transactions are significantly out-of-the-money put options that are insensitive to normal changes in market credit spreads. – A significant change in credit spreads is required to cause a material change in fair value. Credit spread changes did not result in a significant change to fair value losses until the third and particularly the fourth quarters of 2007.
  • 31. 30 AIGFP “Super Senior” CDS Portfolio Total Notional Amounts and Cumulative MTM Loss Type Notional Amount ($ Billions) Cumulative MTM Loss ($ Billions) Corporate Arbitrage $ 70.4 $ 0.2 Regulatory Capital* $ 378.7 $ 0.0 Multi-Sector CDO, of which: High Grade Mezzanine $ 78.2** $ 59.3 $ 18.9 $ 11.3*** $ 7.1 $ 4.2 Total: $ 527.3 $ 11.5*** (December 31, 2007) * Represents Corporate & European Residential Mortgage Regulatory Capital transactions. ** The average amount of defaulted collateral in the multi-sector CDOs is 51 basis points. There are four outliers to this average where defaulted collateral exceeds 10%, the maximum of which is 13.2%. However, in all these cases the subordination is in excess of 43% with the maximum being 64.4%. *** Includes benefit of $310 million attributable to cash flow diversion features.
  • 32. 31 Evolution of AIGFP’s Valuation Methodologies for “Super Senior” Corporate Arbitrage Transactions • At December 31, 2007, AIGFP valued its Corporate Arbitrage “Super Senior” credit derivative transactions using relevant market indices or third party prices. • AIGFP reported a mark-to-market loss in the amount of $226 million under this approach in the fourth quarter of 2007. • At September 30, 2007 AIGFP employed the BET model to value this portfolio, resulting in no noticeable change in fair value.
  • 33. 32 Regulatory Capital (Corporate - Regulatory & European Residential Mortgage) “Super Senior” Transactions Valuation • Transactions entered into are of a highly customized, non-market standard nature, to facilitate regulatory capital relief, rather than for credit risk transfer. • Transactions are expected to terminate in conjunction with the implementation of Basel II (within 12 to 18 months). • AIG conducted a comprehensive analysis of information available at year end, including counterparty motivation, portfolio performance, market place indicators and transaction-specific considerations. • The most compelling market observable data is the termination of $54 billion of transactions in early 2008. AIG was not required to make any payments and was paid a termination fee in some terminations. • Hence AIG believes that these regulatory trades are appropriately valued at zero fair value as of December 31, 2007.
  • 34. 33 Evolution of AIGFP’s Valuation Methodologies/Inputs for “Super Senior” Credit Derivatives Written on CDOs Date September 30, 2007 October 31, 2007 November 30, 2007 (Method A) November 30, 2007 (Method B) December 31, 2007 Methodology • Modified BET • No attribution of Cash Flow Diversion (CFD) • Modified BET • No attribution of Cash Flow Diversion (CFD) • Modified BET • Attribution of Cash Flow Diversion (CFD) using Monte Carlo simulation • Modified BET • Attribution of Cash Flow Diversion (CFD) using Monte Carlo simulation • Negative Basis Adjustment • Modified BET • Attribution of Cash Flow Diversion (CFD) using Monte Carlo simulation • No Negative Basis Adjustment • Overlay of Super Senior Tranche Price Quotes Inputs • Third party credit spreads on generic ABS • Moody’s recovery rates • Third party credit spreads on generic ABS • Third party spreads on RMBS collateral adjusted for relative change in ABX.HE • Moody’s recovery rates • Third party credit spreads on generic ABS • Third party spreads on RMBS collateral adjusted for relative change in ABX.HE • Moody’s recovery rates • Third party prices collected by CDO managers during November for October month- end • Moody’s recovery rates • Third party prices collected by CDO managers during January for December month- end • Moody’s recovery rates
  • 35. 34 Process Followed for December 31, 2007 GAAP Valuation of “Super Senior” CDS Written on CDOs Acquisition &Acquisition & Review of ThirdReview of Third Party Prices ofParty Prices of CollateralCollateral SecuritiesSecurities BenchmarkingBenchmarking to Independentto Independent SourcesSources ModeledModeled SuperSuper SeniorSenior marketmarket value lossvalue loss Overlay ofOverlay of Super SeniorSuper Senior Tranche PriceTranche Price QuotesQuotes Acquisition &Acquisition & Review of ThirdReview of Third Party Prices ofParty Prices of UnderlyingUnderlying Securities ObtainedSecurities Obtained Through CDOThrough CDO Managers:Managers: •• Obtained dealerObtained dealer prices on 70% ofprices on 70% of securities of allsecurities of all portfoliosportfolios combinedcombined •• Derived finalDerived final price byprice by averaging inaveraging in case of multiplecase of multiple quotesquotes •• Reviewed pricesReviewed prices for consistencyfor consistency across ratingsacross ratings and timeand time Benchmarking ofBenchmarking of Third Party Prices toThird Party Prices to Independent PriceIndependent Price Sources:Sources: •• To IDC pricesTo IDC prices (9,180(9,180 securities);securities); •• BloombergBloomberg –– although feweralthough fewer matches (1,124matches (1,124 securities);securities); •• Monthly trendsMonthly trends in ABX.HE indexin ABX.HE index (Series 6(Series 6--1, 61, 6--2,2, 77--1)1) Key InputsKey Inputs to Modifiedto Modified BET ModelBET Model Acquisition andAcquisition and review of other keyreview of other key inputs to theinputs to the Modified BETModified BET model:model: •• WAL ofWAL of securitiessecurities -- Bloomberg;Bloomberg; •• Verification ofVerification of WAL usingWAL using prepaymentprepayment model;model; •• Use of matrixUse of matrix pricing;pricing; •• Diversity score;Diversity score; •• LIBOR curve forLIBOR curve for discounting cashdiscounting cash flows;flows; •• Recovery ratesRecovery rates based onbased on MoodyMoody’’s multis multi-- sector CDOsector CDO recovery datarecovery data Valuation, reviewValuation, review and stress testing ofand stress testing of Modified BETModified BET results of the Superresults of the Super Senior marketSenior market valuation loss:valuation loss: •• Convert price toConvert price to spread;spread; •• Use key inputsUse key inputs to run BET;to run BET; •• Apply OC testsApply OC tests and implementand implement CFD algorithms;CFD algorithms; •• Stress testingStress testing inputs andinputs and validation usingvalidation using separateseparate GaussianGaussian Copula model;Copula model; •• Validation byValidation by applying modelapplying model to observedto observed CDX SS pricingCDX SS pricing Overlaying theOverlaying the Super SeniorSuper Senior Tranche QuotesTranche Quotes Obtained From 12Obtained From 12 Major Dealers to theMajor Dealers to the modified BET modelmodified BET model resultsresults
  • 36. 35 “Super Senior” Multi-Sector CDO Valuation Model • The Binomial Expansion Technique (BET) methodology was originally developed by Moody’s for rating portfolio credit products and is transparent, widely used by market participants and in the public domain. • We modified it to imply default probabilities from market prices for the underlying securities, not from rating agency assumptions. • The model replaces a large collateral pool of correlated assets with a smaller pool of idealized homogeneous, independent assets • The size of the idealized pool, i.e., the number of assets, is given by the Diversity Score. • The BET model placed in a Monte Carlo simulation framework enables AIGFP to: – Derive a loss distribution through time for the portfolio; – Value the important structural features of each transaction.
  • 37. 36 “Super Senior” Multi-Sector CDO Valuation Model Model parameters are derived from independent market sources. • Third party prices for the underlying securities that comprise the collateral of each CDO are obtained from CDO managers. • AIGFP was able to obtain prices for 69% of unique collateral securities comprising the underlying collateral. • From these prices AIGFP derives credit spreads and market-implied default probabilities. • Diversity Scores are generally provided by CDO trustees as the determinant of correlation. • Weighted Average Life for underlying securities are obtained from third-party data providers. • Assumed Recovery Rates for each underlying security is obtained from Moody’s historical experience.
  • 38. 37 “Super Senior” Multi-Sector CDO Valuation Model • The BET model has been reconciled by a separate AIGFP team using a Gaussian Copula model. • Valuation results were consistent when the model’s inputs were calibrated consistently. • Both models confirmed via parameter stress testing that in the current market environment valuations had a relatively low sensitivity to: – the correlation input (diversity score), and – the loss distribution function implied by each model.
  • 39. 38 Negative Basis (1 of 2) • Negative basis refers to the fact that cash and synthetic instruments can often trade at different levels, with cash instruments at a wider spread than CDS. • Cash instruments trade at a spread above the risk-free rate. The spread incorporates the risk of default (credit risk), the market’s price for liquidity risk and, in distressed markets, risk aversion costs. • One component of negative basis is the funding cost. A CDS, i.e., a synthetic position, does not need to be funded initially and hence saves the investor the cost of financing the position and potentially any balance sheet usage. – A typical example: • A cash bond for corporate XYZ trades at a spread of Libor + 250 bps • The credit protection on the same credit to the matching maturity is offered at a spread of Libor +150 bps • In the trade the investor buys both the bond and the credit protection, removing any credit risk embedded in the bond and locking in a positive spread of 100 bps • This funding cost, which is reflected in the market’s price for liquidity risk of cash instruments, is clearly greater in times of crisis or market illiquidity, as the benefit of holding cash increases and the negative basis also tends to rise.
  • 40. 39 Negative Basis (2 of 2) • There is a well established market in corporate credit with investors buying bonds and simultaneously buying matching credit protection on the same instrument in order to “lock in” the positive spread differential. This is termed the negative basis. – There is an active investment grade corporate market to transact, collect market quotes and document its existence. • The negative basis in the structured credit market follows the same principles and exists for the same reasons as in the corporate credit market. – As a result of the ABS market currently being more illiquid, collecting sufficient market quotes and documenting the existence of the negative basis in the structured credit market is problematic. • Our transactions are evidence of an observable negative basis market at inception on CDOs of ABS. • There are other components which also drive the basis, some of which make it less negative or may even make it positive at times.
  • 41. 40 Why Did We Not Incorporate Negative Basis into Our Valuations as of December 31, 2007? • AIGFP did not adjust its fair valuations for the negative basis, even though it believes such a difference exists. • AIG must collect sufficient observable evidence that supports existence and realization of negative basis to permit AIG to factor negative basis into its valuation methodology. • AIG is not able to obtain sufficient objective evidence at this time as a result of the market dislocation and lack of trading.
  • 42. 41 Similarities Between Excess Casualty Insurance & AIGFP’s “Super Senior” Credit Derivative Portfolio on Multi-Sector CDOs • AIG acts as principal • AIG retains underwriting control • Relatively high attachment points • Gives AIG access to specialized (non-commodity) markets • AIG can “pick and choose” risks – broker has no binding authority • Liabilities are generally not traded, but held to settlement • Reinsurance used selectively to take advantage of market pricing Excess Casualty Insurance “Super Senior” Credit Derivative Portfolio on Multi-Sector CDO • AIG acts as principal • AIG retains underwriting control • Relatively high attachment points • Gives AIG access to specialized (non- commodity) markets • Largely bespoke transactions with tailored contractual terms • “Super Senior” credit derivatives are generally not traded, but held to settlement • Protection purchased selectively for rated layers on portfolios of reference obligations
  • 43. 42 Valuation Principles Applied By AIG • “The price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date” • “NOT a forced transaction (for example, a forced liquidation or distressed sale)”; but • Represents current “exit price” in the principal market “Fair Value” Under U.S. GAAP “Market Consistent Settlement Value” under AIG’s Economic Capital Model (ECM) • For tradeable assets and liabilities, represents the market value without adjustment for “own credit standing” • For non-tradeable assets and liabilities (e.g., those requiring use of “Level 3” inputs under FAS 157), represents current estimated value of asset or liability plus a risk margin for bearing risk out to settlement • No “exit” assumed at valuation date – avoids imputing characteristics of a liquidation in times of market stress Source: Centre for Audit Quality (CAQ) , White Paper, “Measurements of Fair Value in Illiquid (or less liquid) Markets”; October 3, 2007 (www.aicpa.org)
  • 44. 43 Illustration of Key Differences Between Fair Value & Market Consistent Settlement Value for Non-Tradeable Assets/Liabilities • Most significant difference is in risk and service margin component • Under Market Consistent Settlement Value, an allowance is made for the cost of bearing risk (only) until settlement of the liabilities using a market standard technique (e.g., Cost of Capital) • Under current exit value, the “risk & service margin” may be materially inflated in times of market crisis for the opportunity of a buyer of the liabilities to recover a super normal return for : – Liquidity Risks – Sunk Costs & Opportunity Costs – Return on Franchise Value – Supply/Demand Imbalance Advantages Example: Valuation of Non-Tradeable Liability Consequent to Liability Market Crisis (e.g. Post KRW in P&C Insurance) Fair Value Based on Current Exit Prices (e.g. per FAS 157) Market Consistent Settlement Value Risk & Service Margin Remove Own Credit Risk Adjustment Current Discounted Estimate of Contractual Cash Flows Market Consistent Settlement Value is more appropriate for determining the economic position of AIG, as AIG generally intends to, or is obligated to, hold its illiquid asset and liability positions until maturity. -ILLUSTRATION -
  • 45. 44 Determination of Market Consistent Settlement Value of AIGFP’s “Super Senior” Credit Derivative Portfolio on Multi-Sector CDOs for AIG’s Economic Capital Model (December 31, 2007) 1.71 0.81 0.90 0.00 0.50 1.00 1.50 2.00 2.50 Estimated Realizable Loss Consequent to Severe Stress Cost of Capital Risk Margin Market Consistent Settlement Value Loss Under AIG's ECM AIG’s conservative estimate of Market Consistent Settlement Value Loss for Determining Available Economic Capital: 1) Realizable losses associated with the “severe” stress scenario over-ride best estimate loss (allows for possible change in credit risk fundamentals) 2) A risk margin is added to the component under 1), allowing for the cost of servicing capital requirements* 3) Market Consistent Settlement Value is the sum of 1) and 2) above Conservative Estimate of Market Consistent Settlement Value Loss Under AIG’s Economic Capital Model (ECM) * Determined using the Cost of Capital approach recommended by the CRO Forum for non-tradeable/ non- hedgeable risks (refer to www.croforum.org) $ BN$ BN
  • 46. 45 Adjustments Made to Unrealized GAAP Market Valuation Loss to Determine Available Economic Capital Under AIG’s ECM 1.71 9.54 6.20 11.25 0.00 2.00 4.00 6.00 8.00 10.00 12.00 Unrealized Market Valuation Loss Carried on GAAP Balance Sheet Market Consistent Settlement Value Loss Pre-Tax Adjustment to Unrealized Losses After-Tax Adjustment to GAAP Equity • At June 30, 2007 AIG used GAAP equity as a conservative proxy for Available Economic Capital • For December 31, 2007: – AIG will use Market Consistent Embedded Value* as its estimate of Available Economic Capital for the Life & Retirement Services segment – For the General Insurance segment, a consistent approach will be used – These valuation approaches are consistent with the market consistent settlement value approach AIG has applied to FP’s Super Senior credit derivative portfolio of Multi-Sector CDOs Market Consistent Settlement Value Adjustment to Determine Available Economic Capital (December 31, 2007) $BN * Currently being independently reviewed and certified by Towers Perrin. For the purposes of determining Available Economic Capital, AIG believes it is reasonable to make a positive market consistent settlement value adjustment of $6.2 billion in respect of the AIGFP Unrealized Loss to its GAAP Reported Total Shareholders’ Equity as at December 31, 2007.
  • 47. 46 Outline • Business Rationale, Portfolio Composition & Underwriting Standards • Fundamental Risk Assessment & Stress Testing • Accounting, Valuation Fundamentals & Economic Capital • Conclusions & Next Steps
  • 48. 47 Conclusions • AIGFP “Super Senior” credit derivative business was underwritten to an AIGFP super senior standard which projects zero losses at inception. • There has been deterioration in the credit quality and the market’s assessment of expected losses in the underlying collateral securities and AIGFP’s “Super Senior” credit derivative portfolios. • In accordance with GAAP, AIG recognized a sizable unrealized market valuation loss in 2007, consequent to severe market disruption and credit deterioration, particularly in US sub-prime mortgages, occurring predominantly in the fourth quarter. This market adjustment represented management’s best estimate of the exit value of this portfolio into the current illiquid and distressed market. This volatile market may persist for some time. • Despite the unrealized market valuation loss: – Based upon its most current analyses, AIG believes that any losses AIGFP may realize over time as a result of meeting its obligations under these derivatives will not be material to AIG’s consolidated financial position, although it is possible that such realized losses could be material to AIG’s consolidated results for an individual reporting period. – Except to the extent of credit impairment losses, AIG expects the unrealized market valuation losses to reverse over the remaining life of the portfolio. • AIGFP wrote credit protection as a principal based upon sound underwriting procedures and has the ability and intent to hold its positions until contract maturity or call by the counterparty.
  • 49. 48 Next Steps • Remediate the material weakness in internal control over financial reporting with respect to the controls over the AIGFP “Super Senior” credit derivative portfolio valuation process and oversight thereof. • Continue to monitor closely the portfolio risk of realized loss through fundamental analysis, modeling and stress testing. • Enhance current valuation methodologies and processes – Improve the timeliness and comprehensiveness of data inputs – Consider/develop/implement additional modeling techniques – Continue to observe the market • Pursue opportunities in the market.
  • 51. 50 AIG Insurance Investment Portfolios • The investment portfolios of AIG’s insurance companies are managed by AIG Investments (AIGI)* on their behalf. • These portfolios are managed on a spread investment or Asset-Liability Management model, not as a transactional business. As a result, we do not: – “Warehouse” residential mortgage loans or securitizations; or – Retain residual or other securities from residential mortgage backed securities (RMBS) activities. • AIGI’s RMBS and commercial mortgage backed securities (CMBS) are predominantly held as “Available for Sale” securities, not as trading positions. Hence, our underwriting focus is on ultimate collectibility, not short-term market movements. • AIG, as with all investments, purchases RMBS, Asset-Backed Commercial Paper, Structured Investment Vehicles, and RMBS-based Collateralized Debt Obligations (CDOs) based on proprietary internal research. • All information presented in this document is as of December 31, 2007, unless otherwise noted. • All figures are based on amortized cost** unless otherwise indicated. • Ratings used in this presentation are external ratings, or equivalent, based on AIG’s internal risk rating process. * For purposes of this presentation, AIGI is used to denote AIG Insurance Investment Portfolios. ** Amortized cost is the cost of a debt security adjusted for amortized premium or discount less other-than-temporary impairments.
  • 52. 51 AIG Insurance Investment Portfolios Worldwide Insurance and Asset Management Bond Portfolios • AIGI’s bond portfolios* had a fair value of $491.3 billion at December 31, 2007. • The securities are highly rated (approximately 95% are investment grade). • The bond portfolios are also well-diversified geographically. * Fixed Maturities: Bonds available for sale, Bonds held to maturity, Bonds trading securities and Bonds available for sale included in Securities lending collateral $286.7 Billion $204.6 Billion Foreign Bonds by Ratings Domestic Bonds by Ratings AA 44% A 27% AAA 21% Lower 2% BBB 6%BBB 15% Lower 7% AAA 51% A 12% AA 15%
  • 53. 52 AIG Insurance Investment Portfolios Domestic Bonds by Category Other ABS, 2% CDO Debt, 1% Credit, 42% CMBS, 4% RMBS, 27% U.S. Government, 1% Municipal, 23% $286.7 Billion
  • 54. 53 AIG Insurance Investment Portfolios Accounting and Valuation • AIGI accounts for its RMBS, CMBS and CDOs in accordance with FAS 115, FSP FAS 115-1 and EITF 99-20. – These securities are predominantly classified as available for sale securities under FAS 115. – Changes in fair value are reported in other comprehensive income, net of tax, as a component of shareholders’ equity until realized. – Realization through earnings occurs when the position is either sold or is determined to be other-than-temporarily impaired. • AIGI utilizes external pricing vendors as a primary pricing source. – 95% of the portfolio fair values are derived from prices provided by industry standard commercial pricing vendors – such as IDC, Bloomberg and Lehman Brothers. – Vendor pricing methodology and broker prices are internally reviewed for reasonableness, but these securities are not valued solely based on internal models. • The valuation of these securities varies by the type of collateral, the position in the capital structure and the vintage.
  • 55. 54 AIG Insurance Investment Portfolios Other Than Temporary Impairments (OTTI) • AIG’s senior management evaluates its investments for impairment such that a security is considered a candidate for other-than-temporary impairment if it meets any of the following criteria: – Trading at a significant (25 percent or more) discount to par, amortized cost (if lower) or cost for an extended period of time (nine consecutive months or longer); – The occurrence of a discrete credit event resulting in (i) the issuer defaulting on a material outstanding obligation; (ii) the issuer seeking protection from creditors under the bankruptcy laws or any similar laws intended for court supervised reorganization of insolvent enterprises; or (iii) the issuer proposing a voluntary reorganization pursuant to which creditors are asked to exchange their claims for cash or securities having a fair value substantially lower than par value of their claims; or – AIG may not realize a full recovery on its investment, regardless of the occurrence of one of the foregoing events. • An impairment charge may also be taken in light of a rapid and severe market valuation decline because AIG could not reasonably assert that the recovery period would be temporary (severity losses). • AIG Investment’s Chief Investment Officer – Insurance Companies and Chief Credit Officer make credit- related OTTI recommendations using three categories: a) likely to recover; b) possible to recover; and c) unlikely to recover, based on a detailed written description of the circumstances of each security. • In addition, in accordance with EITF 99-20 an analysis of the anticipated cash flows supporting each asset backed security (ABS), representing rights to receive cash flows from asset pools, such as CDOs, RMBS, CMBS, etc., and generally rated below AA-, is prepared and reviewed for impairment. • All credit-related OTTI recommendations, together with supporting documentation, are reviewed on a quarterly basis and approved by AIG’s Chief Credit Officer. The AIG Chief Credit Officer must also determine whether there are any additional securities (not on the list submitted by AIG Investment’s Chief Investment Officer – Insurance Companies) that should be written down.
  • 56. 55 AIG Insurance Investment Portfolios Consolidated Summary of Gains & Losses Financial Effect of Market Disruption For the Year Ended December 31, 2007 Realized and Unrealized Gains / Losses (Pre-tax) ($ Million) Total AIG* Amount Attributable to RMBS Portfolio Net realized capital gains (losses) ($3,592) ($1,647) of which, Sales Activity $1,237 ($30) OTTI ($4,072) ($1,617) Other** ($757) ($0) Unrealized (depreciation) appreciation of investments (included in Other comprehensive income) ($8,046) ($5,070) of which, AAA-rated RMBS (depreciation) ($4,633) ($4,633) AA-rated RMBS (depreciation) ($752) ($752) Lower than AA-rated RMBS (depreciation) ($118) ($118) RMBS appreciation $433 $433 • The other-than-temporary impairments and unrealized losses result primarily from the severe credit and liquidity market turmoil. * Excludes AIGFP’s super senior credit default swap portfolio. **Consists predominantly of foreign exchange related losses.
  • 57. 56 AIG Insurance Investment Portfolios RMBS Portfolios
  • 58. 57 AIG Insurance Investment Portfolios RMBS Overview • Non-agency RMBS are issued in tranche structures, such that the lower tranches absorb losses on the underlying collateral in the pool and thus insulate the higher rated tranches from loss. – The structure and size of each tranche depend on the nature of the collateral, rating agency analysis and models of default scenarios. – As a general rule, AAA and AA rated securities are structured to withstand default losses within the collateral that are multiples of historical norms without any loss of principal or interest. • Holdings of global residential mortgage market products total approximately $89.9 billion at December 31, 2007, or about 10% of AIG’s total invested assets. • Within AIGI’s $75.3 billion non-agency portfolio, about 89% are AAA-rated and 8% are AA-rated. – Holdings rated BBB or below total approximately $0.5 billion, under 1% of the portfolio and less than 0.1% of total invested assets. – About $7.4 billion (9.9%) of the $75.3 billion is “wrapped” by monoline insurance. – Approximately $2.1 billion of principal was paid down during the fourth quarter. RMBS Type Amortized Cost ($ Billion) % Fair Value ($ Billion) % Agency Pass-Through and CMO Issuances $14.6 16% $ 14.8 17% Prime Non-Agency (incl Foreign and Jumbo MBS related securities) 21.6 24% 21.1 25% Alt-A RMBS 25.3 28% 23.8 28% Subprime RMBS 24.1 27% 21.2 25% Other Housing-Related Paper 4.3 5% 3.9 5% Total RMBS $89.9 100% $84.8 100%
  • 59. 58 AIG Insurance Investment Portfolios RMBS Portfolio Amortized Cost ($ Millions) RATING HOLDINGS AGENCY AAA AA A BBB NON INV TOTAL AGENCY $14,575 $ - $ - $ - $ - $ - $14,575 PRIME JUMBO - 12,708 1,794 483 143 1 15,129 ALT-A - 23,967 994 309 71 8 25,349 SUBPRIME - 20,843 2,833 388 10 - 24,074 SECOND-LIEN - 2,055 40 85 40 3 2,223 HELOC - 1,989 - - - - 1,989 FOREIGN MBS - 5,735 174 294 220 - 6,423 OTHER - 35 13 27 14 - 89 TOTAL $14,575 $67,332 $5,848 $1,586 $498 $ 12 $89,851
  • 60. 59 AIG Insurance Investment Portfolios Total RMBS - $89.9 Billion AAA trancheAAA tranche AA trancheAA tranche A trancheA tranche BBB trancheBBB tranche AA $5.9 Billion (6.5%) A $1.5 Billion (1.7%) BBB $0.5 Billion (0.6%) Equity <$0.1 Billion (0.1%) AAA & Agency $81.9 Billion (91.1%) RMBS (Collateral pool of residential mortgages) RMBS (Collateral pool of residential mortgages) BB and lower Equity tranche BB and lower Equity tranche Payment Waterfall (principal + interest) Priority First Last
  • 61. 60 AIG Insurance Investment Portfolios Total RMBS Exposure by Vintage - $89.9 Billion $Billions 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 AAA 3.95 6.26 7.14 16.75 27.53 20.27 AA 0.07 0.84 0.62 1.24 2.52 0.56 A 0.03 0.26 0.36 0.52 0.28 0.07 BBB 0.00 0.06 0.08 0.22 0.11 0.11 Prior 2003 2004 2005 2006 2007 AIGI focuses almost exclusively on AAA and AA rated investments with relatively short tenors Vintage Weighted average expected life (WAL) is 5.09 years
  • 62. 61 AIG Insurance Investment Portfolios Subprime RMBS • Through its initial investment analysis, combined with ongoing monitoring, virtually all of AIGI’s 2006/2007 exposure is high up in the capital structure (AAAs and AAs). • Nevertheless, lifetime loss estimates for subprime securities – currently 15% - 21% for the 2006 vintage – have been increasing. • Generic AA and AAA rated securities can withstand cumulative loss percentages ranging from the high teens to the mid-to-high 20s. In effect, AIGI is exposed to significant downgrade risk and market price volatility, but AIGI believes that the risk of ultimate loss is not expected to be significant. • Recent Fed rate cuts have been a modest loss mitigant: – The decline in short-term rates increases excess interest, which is used to absorb losses. – Payment shock for ARM borrowers at interest re-set has been significantly reduced. 2006 Vintage Credit Enhancement for AIGI* Rating Original Credit Enhancement Current Credit Enhancement AAA 20.7% 29.6% AA+ and lower 15.5% 21.5% *Source: Intex
  • 63. 62 AIG Insurance Investment Portfolios Subprime RMBS - $24.1 Billion AAA trancheAAA tranche AA trancheAA tranche A trancheA tranche BBB trancheBBB tranche AA $2.8 Billion (11.6%) A $0.4 Billion (1.7%) BBB $10 Million (0.0%) Equity <$0.1 Million (0.0%) AAA $20.9 Billion (86.7%) RMBS (Collateral pool of residential mortgages) RMBS (Collateral pool of residential mortgages) BB and lower Equity tranche BB and lower Equity tranche Last Payment Waterfall (principal + interest) Priority First
  • 64. 63 AIG Insurance Investment Portfolios Subprime RMBS Exposure by Vintage - $24.1 Billion $Billions AIGI focuses almost exclusively on AAA and AA rated investments with relatively short tenors Vintage Weighted average expected life (WAL) is 4.12 years 0.0 2.0 4.0 6.0 8.0 10.0 12.0 AAA 0.14 0.40 0.60 5.81 9.11 4.79 AA 0.01 0.04 0.13 0.33 1.93 0.40 A 0.01 0.08 0.10 0.13 0.04 0.03 BBB - - - 0.01 - - Prior 2003 2004 2005 2006 2007
  • 65. 64 AIG Insurance Investment Portfolios Alt-A RMBS • In the Alt-A sector, AIGI’s initial investment analysis and credit selection process have resulted in most of our exposure being at the super senior* AAA level, especially in the 2006/2007 vintages. • Over 98% of our Alt-A exposure is rated AAA or AA. • While Alt-A loss expectations have climbed into the 2.25 – 3.25% range,** AIGI’s investment decisions with respect to this portfolio have reduced portfolio risk in the current downturn. • Consequently, AIGI is exposed to limited downgrade risk and loss of investment principal in its Alt-A portfolio. *A super senior AAA is structured with credit enhancement in excess of that required by the rating agencies at the AAA level **Source: Lehman Brothers, January 2008, Global Relative Value Conference ***Source: Intex 2006 Vintage Credit Enhancement for AIGI*** Rating Original Credit Enhancement Current Credit Enhancement AAA 19.0% 21.3% AA+ and lower 4.9% 6.5%
  • 66. 65 AIG Insurance Investment Portfolios ALT-A RMBS -$25.3 Billion AAA trancheAAA tranche AA trancheAA tranche A trancheA tranche BBB trancheBBB tranche AA $1.0 Billion (4.0%) A $0.3 Billion (1.2%) BBB $71 Million (0.3%) Equity $8 Million (0.0%) AAA $23.9 Billion (94.5%) RMBS (Collateral pool of residential mortgages) RMBS (Collateral pool of residential mortgages) Payment Waterfall (principal + interest) Priority First BB and lower Equity tranche BB and lower Equity tranche Last
  • 67. 66 AIG Insurance Investment Portfolios ALT-A RMBS Exposure by Vintage - $25.3 Billion $Billions A A AAAAAA AA AA AAAA AA 0.0 2.0 4.0 6.0 8.0 10.0 12.0 AAA 0.21 0.63 0.98 5.15 10.08 6.92 AA 0.03 0.21 0.16 0.44 0.13 0.01 A - 0.04 0.06 0.15 0.05 0.01 BBB - 0.01 0.02 0.04 0.01 - Prior 2003 2004 2005 2006 2007 AIGI focuses almost exclusively on AAA and AA rated investments with relatively short tenors Vintage Weighted average expected life (WAL) is 4.04 years
  • 68. 67 AIG Insurance Investment Portfolios Prime Jumbo RMBS • The domestic prime jumbo RMBS portfolio totaled $15.1 billion at December 31, 2007. • Delinquencies and defaults have generally been well controlled in the prime jumbo market. • Although AIGI took more credit risk in this sector, the majority of our exposure (96%) continues to be in AAAs and AAs. • Loss expectations have climbed but in general remain below the BBB attachment point (approximately 0.5%). • The weighted average expected life of the portfolio is 6.49 years. 2006 Vintage Credit Enhancement for AIGI* Rating Original Credit Enhancement Current Credit Enhancement AAA 8.5% 9.7% AA+ and lower 1.7% 2.1% *Source: Intex
  • 69. 68 AIG Insurance Investment Portfolios RMBS Rating Agency Actions* *Based on 1st Agency to downgrade or put on watch – If on downgrade list, not included on watch list. Source: Moody’s Investors Service, Standard & Poor’s, and Fitch. **January 1, 2008 through February 25, 2008 • Negative rating actions accelerated in the fourth quarter of 2007 and the first quarter of 2008. • The brunt of these actions was borne by the subprime, Alt-A and 2nd lien sectors. The prime jumbo sector was relatively untouched. • An overwhelming majority of the securities on “Watch List Negative” are rated AAA or AA. • Currently we expect very few of the affected securities to ultimately incur a loss to investment principal. Fourth Quarter 2007 YTD 2008** Number of Securities Value ($ Million) Number of Securitie s Value ($ Million) % of Non- Agency RMBS Portfolio Downgraded 35 $443 128 $3,578 4.8% Watch List Negative 96 $3,396 252 $9,694 12.9% Upgrades 0 $0 4 $6 0.0%
  • 70. 69 AIG Insurance Investment Portfolios CMBS Portfolios
  • 71. 70 AIG Insurance Investment Portfolios CMBS Portfolios Overview • AIGI’s CMBS portfolio demonstrated strong credit performance in 2007: – The upgrade / downgrade ratio is significantly better than for the U.S. CMBS universe; and – Delinquencies in underlying collateral are very low and are only approximately 50% of those for the general U.S. CMBS universe. • Approximately 9% of the CMBS exposure (by amortized cost) comes from the re- securitization of CMBS and commercial real estate (CRE) CDOs. Two-thirds of the loans underlying these securities are seasoned 25 months or more. • Our initial investment analysis, combined with ongoing surveillance, has helped protect the CMBS portfolio from the more aggressive underwriting standards that crept into the commercial loan market in 2007. • The portfolio is performing in accordance with expectations from a credit perspective. – For the year ended December 31, 2007, other-than-temporary impairments totaled $147 million, and were recognized primarily as a result of severity of price declines, not credit events. Net unrealized losses were $919 million, about 3.8% of the total portfolio. – No actual credit-related losses to investment principal have been incurred to date.
  • 72. 71 AIG Insurance Investment Portfolios CMBS Portfolios Description Amortized Cost ($ Million) % CMBS (traditional) $21,003 87.81% ReREMIC/ CRE CDO 2,221 9.29% Agency 347 1.45% Other 347 1.45% TOTAL $23,918 100.00% • AIGI’s CMBS portfolio is predominantly comprised of traditional commercial mortgage backed securities. • The underlying collateral of the portfolio is of high quality with close to 89% rated AAA / AA and approximately 99% investment grade. Ratings N/A 1.0% AA 10.9%A 8.7% AAA 78.0% BBB 1.2% BIG 0.2%
  • 73. 72 AIG Insurance Investment Portfolios CMBS Portfolios Vintage % 2007 22.0% 2006 15.5% 2005 18.3% 2004 15.1% 2003 4.9% 2003 & Older 24.2% 100.0% Top 10 States Amortized Cost ($ Million) % NY $3,640 15.2% CA 3,140 13.1% TX 1,476 6.2% FL 1,322 5.5% VA 773 3.2% IL 723 3.0% NJ 696 2.9% PA 588 2.5% GA 576 2.4% MA 555 2.3% $13,489 56.3% • The CMBS portfolio is well-diversified geographically, with the top ten states representing slightly over 56% of total exposure. • The majority of the portfolio is of older vintages, although about 22% is from the 2007 vintage.
  • 74. 73 AIG Insurance Investment Portfolios CMBS Portfolios – ReREMIC/ CRE CDO Holdings Rating Amortized Cost ($ Million) % AAA/Aaa $1,013 45.6% AA/Aa 229 10.3% A/A 826 37.2% BBB/Baa 131 5.9% BB/Ba 22 1.0% $2,221 100.0% Top 10 States % CA 15.2% NY 10.2% TX 7.2% FL 5.7% VA 3.6% IL 3.6% PA 3.1% NJ 2.9% GA 2.8% MA 2.7% 57.0% • Only 9.3% of the total CMBS portfolio is represented by ReREMIC / CRE CDO securities. – Close to 58% of these securities are ReREMIC and the other 42% are CRE CDOs – Highly rated collateral (99% investment grade rated) – Geographically diversified – Well-seasoned portfolio with over 65% of the underlying loans seasoned over 25 months Collateral Seasoning (mos) 37+ 45.1% </= 12 3.5% 13 – 24 31.1% 25-36 20.3%
  • 75. 74 AIG Insurance Investment Portfolios CMBS Portfolios Current Delinquency % Source: Trepp, Morgan Stanley and Intex Delinquencies as of 2/4/08 • The U.S. CMBS sector has enjoyed strong credit performance. Delinquencies are near historic lows and have remained below 1% since 2005. • AIGI’s CMBS portfolio is currently outperforming the U.S. CMBS universe. 0.04 0.02 0.13 0.12 0.06 0.02 0.27 0.07 0.50 0.23 0.00 0.10 0.20 0.30 0.40 0.50 0.60 60 90+ FCL RE O Total US CM B S Univers e A IG CM B S P ortfolio
  • 76. 75 AIG Insurance Investment Portfolios CMBS Portfolios Rating Actions Full Year 2007 AIGI CMBS Portfolio U.S. CMBS Universe Combined 43.2:1 8.4:1 Investment Grade Bonds 43.2:1 21.4:1 Below Investment Grade Bonds n/a 1.8:1 • AIGI’s CMBS upgrade / downgrade ratio is significantly better than that of the U.S. CMBS universe.
  • 77. 76 AIG Insurance Investment Portfolios CDOs
  • 78. 77 AIG Insurance Investment Portfolios CDO Overview • Investments in collateralized debt obligations (CDOs) total $4.55 billion representing approximately 1% of total insurance investment portfolios. – Only $58 million are mezzanine ABS CDOs with subprime exposure. – Only 0.4% of total CDO holdings were downgraded in 2007. • CDO pricing has been adversely affected by current market conditions. • Based on our current analysis, the risk of ultimate loss of investment principal is low.
  • 79. 78 AIG Insurance Investment Portfolios CDO Overview • As of December 31, 2007 the composition of the $4.55 billion CDO global portfolio is as follows: • The weighted average market price of the CDO portfolio was $81(2) as of December 31, 2007, down from $88(2) as of September 30, 2007. (1) Below Investment Grade (2) As compared to par of $100 Ratings ($ in Billions) Amortized Cost % AAA $0.80 17.6% AA 1.02 22.5% A 2.21 48.5% BBB 0.42 9.2% BIG(1) and Equity 0.10 2.2% Total $4.55 100% Collateral Type ($ in Billions) Amortized Cost % Bank Loans (CLO) $2.14 47.0% Synthetic Investment Grade 1.56 34.3% Other 0.79 17.4% Subprime ABS 0.06 1.3% Total $4.55 100%
  • 80. 79 AIG Insurance Investment Portfolios Bank Loan CLOs • The composition of the $2.14 billion in collateralized loan obligations (CLO) holdings (47% of the total CDO portfolio) by rating is as follows: • Holdings continue to exhibit good performance, reflecting the relatively low default rate environment for leveraged loans. We expect the default rate to increase, but to still moderate levels. • None of the debt tranches in our CLO portfolio has been downgraded or placed on negative watch by the rating agencies. • None of our tranches is deferring interest. • 80% of CLO holdings are rated A or better. Ratings Amortized ($ in Billions) Cost % AAA $0.02 1% AA 0.13 6% A 1.55 73% BBB 0.35 16% BIG and Equity 0.09 4% Total $2.14 100%
  • 81. 80 AIG Insurance Investment Portfolios Synthetic Investment Grade Rated Corporate CDOs • The $1.56 billion in Investment Grade Corporate Synthetic holdings (34% of the total CDO portfolio) is highly rated as shown below: • The underlying portfolios are predominantly CDS against investment grade corporate bonds and loans. • Holdings continue to exhibit good performance, reflecting the relatively benign default environment for investment grade rated corporates. The initial credit enhancement levels remain intact. • None of our tranches has been downgraded or placed on negative watch by the rating agencies. • None of our tranches is deferring interest. • 76% of these holdings are rated AA or better and 99% are rated A or better. Ratings ($ in Billions) Amortized Cost % AAA $0.44 28% AA 0.74 48% A 0.37 23% BBB 0.01 1% Total $1.56 100%
  • 82. 81 • 91% of the remaining CDO holdings, which include primarily market value and older vintage CDOs, is rated A or better. • These holdings of $793 million have maintained their value ($785 million fair value). • These CDO holdings are experiencing stable performance. • The composition of the holdings from a ratings standpoint is as follows: AIG Insurance Investment Portfolios Other CDOs BIG and Equity 2% BBB 7% AAA 41% AA 17% A 33%
  • 83. 82 AIG Insurance Investment Portfolios Monoline Exposure
  • 84. 83 AIG Insurance Investment Portfolios Monoline Exposure • AIGI’s exposure to monolines totals $42.2 billion, 99% of which are financial guarantees. – 74% of the total exposure relates to municipal bonds, which are highly rated, even without the financial guarantees. – AIGI does not rely primarily on financial guarantees with regards to making investment decisions in its bond portfolio. • The composition by asset class is as follows: 1) Refers to cash collateral accounts in certain synthetic CDOs. $399 million of this exposure is investment agreements with financial guarantee insurance policies provided by the monolines (includes $220 million of fully collateralized investment agreements and $179 million of investment agreements which are subject to collateral posting requirements, should the monoline guarantor be downgraded). Also includes $41 million in an investment agreement issued by a monoline with a corporate guarantee provided by a highly rated non-monoline guarantor. (2) Represents amortized cost and fair value related to $57 million of bonds and $136 million notional of CDS. (3) The fair value for the bond portion is $52 million and the market value for the CDS portion is ($13) million. Asset Class Amortized Fair ($ in Billions) Cost Value Municipals $31.41 $31.88 RMBS/CMBS 7.40 6.82 ABS 2.14 2.01 Corporates 0.76 0.80 Investment Agreements in CDOs (1) 0.44 0.36 Total Insured 42.15 41.87 Direct Corporate Exposure (2),(3) 0.06 0.04 Total Exposure $42.21 $41.91
  • 85. 84 AIG Insurance Investment Portfolios Monoline Exposure • Exposure by monoline entity and by type of exposure is as follows: (1) Amounts above are exclusive of $136 million in Notional of CDS as follows: $56 million (AMBAC), $29 million (MBIA), and $51 million (Assured Guaranty). (2) Other includes the amortized cost of corporate exposure and Investment Agreements in CDOs. Financial Guarantees Amortized Cost ($ in Billions) (1) Other (2) MBIA $13.10 $0.16 FSA 10.45 0.14 AMBAC 9.13 0.15 FGIC 7.37 0.04 SCA (XLCA/XLFA) 0.76 - CIFG 0.67 - Assured Guaranty Corp. 0.22 - Multiple 0.01 - $41.71 $0.49
  • 86. 85 AIG Insurance Investment Portfolios Monoline Exposure by Underlying Ratings • The quality of the insured portfolio is high, with 84% having ratings of A or above. This is also a reflection of the large percentage of municipals in our wrapped monoline portfolio. (1) Includes RMBS/CMBS and ABS underlying ratings, which are based solely on AIG’s internal ratings assessment. (2) Excludes $440 million of Investment Agreements in CDOs and $57 million of corporate exposure. Underlying Ratings (1) Amortized Cost ($ in Billions) (2) % of Total AAA $2.90 7.0% AA 21.35 51.1% A 10.89 26.1% BBB 3.62 8.7% BB 2.40 5.8% B 0.04 0.1% CCC 0.45 1.1% Non Rated 0.06 0.1% $41.71 100.0%
  • 87. 86 AIG Insurance Investment Portfolios Monoline Exposure Municipal Bonds by Underlying Ratings • As of December 31, 2007 AIGI’s municipal bond portfolio totaled $61.5 billion. • More than 99% of the total municipal bond portfolio is rated A or better, even without the financial guarantees. • More than 98% of AIGI’s insured municipal bond portfolio is rated A or better, even without the financial guarantees. Insured Portfolio ($31.4 Billion) No Underlying 0.1% Baa/BBB 1.2% AAA .4% Aa/AA 65.9% A 32.4% Total Portfolio ($61.5 Billion) Aa/AA 53.1% Aaa/AAA 14.6% AAA Pre-re/ETM 12.2%A 19.4% No Underlying 0.1% Baa/BBB 0.7%
  • 88. 87 AIG Insurance Investment Portfolios Monoline Exposure RMBS/CMBS • AIGI’s insured RMBS/CMBS portfolio is diversified across product types. • 63% of the underlying RMBS/CMBS portfolio is internally rated investment grade. • Below Investment Grade represents primarily second lien and HELOC pools that have experienced worse than anticipated performance. AIG Internal Ratings Amortized ($ in Billions) Cost % AAA $2.58 35% AA 0.32 4% A 0.42 6% BBB 1.31 18% BB 2.32 31% CCC 0.45 6% $7.40 100% Asset Class Amortized ($ in Billions) Cost % SECOND LIEN $1.96 26% HELOC 1.93 26% ALT-A 1.69 23% Subprime 1.26 17% JUMBO 0.42 6% Foreign MBS 0.06 1% CMBS 0.05 1% Manufactured Housing 0.03 0% $7.40 100%
  • 89. 88 AIG Insurance Investment Portfolios Monoline Exposure ABS • AIGI’s insured non-RMBS/CMBS ABS portfolio is diversified across product types. • 98% of the underlying ABS portfolio is internally rated investment grade. • We do not rely on monoline support as the primary source of repayment and thus do not look to it as a material consideration to the collectibility of the portfolio. Asset Class Amortized ($ in Billions) Cost % Business/Franchise Loan $0.55 26% Auto Loan 0.48 22% Future Flow 0.42 19% Lot Loan 0.26 12% Project Finance & Other 0.19 9% Railcar Loan/Lease 0.10 5% Timeshare 0.10 5% Credit Card 0.04 2% $2.14 100% AIG Internal Ratings Amortized ( $ in Billions) Cost % AAA $0.20 9% AA 0.16 7% A 0.25 12% BBB 1.49 70% BB 0.04 2% CCC 0.00 0% $2.14 100%
  • 90. 89 AIG Insurance Investment Portfolios Monoline Exposure • Sound fundamental credit underwriting is the foundation for all of AIGI’s investment decisions, and financial guarantees are viewed as a secondary form of payment for all municipal and other wrapped investments. – More than 99% of the underlying $31.4 billion insured municipal bond portfolio is rated investment grade even without the financial guarantees. – 63% of the underlying $7.4 billion insured RMBS/CMBS portfolio is internally rated investment grade. – 98% of the underlying $2.1 billion insured ABS portfolio is internally rated investment grade. • Currently there are 10 RMBS Second Lien, Home Equity and Subprime transactions totaling $380 million, or less than 1% of AIGI’s total insured portfolio, that are known to be receiving contractual payments through their financial guarantees.
  • 91. 90 AIG Insurance Investment Portfolios: Conclusion
  • 92. 91 AIG Insurance Investment Portfolios In Conclusion • The global financial markets remain under considerable stress with reduced financing opportunities for residential mortgage borrowers. • Market expectations for losses in the 2006/2007 U.S. subprime and Alt-A vintages have increased substantially. • Our preference for exposures higher in the capital structure limits our risk of investment principal loss, yet our subprime holdings remain exposed to material market price volatility and downgrade risk. • Our Alt-A holdings continue to benefit from our decision to require credit-enhanced levels in deals beyond the AAA enhancement required by the rating agencies. • Mark-to-market write-downs have increased, but ultimate investment principal losses are not expected to be significant. • Though our CMBS holdings have suffered from volatile market pricing, underlying fundamentals remain strong, with very low delinquencies and a high upgrade / downgrade ratio. • Our CDO holdings have suffered minimal downgrades in 2007, and we have little exposure to the currently struggling subprime CDO market. • Excluding municipal bonds, AIGI’s exposure to monoline insurers is a moderate percentage of our bond holdings. None of these holdings relied on financial guarantees as a primary source of repayment at the time of acquisition. • We view monoline insurance as a secondary consideration to the credit-worthiness of the municipal bond portfolio.
  • 94. 93 United Guaranty (UGC) • UGC, as a broad market participant, operates in an inherently cyclical business that is highly correlated to the fortunes of the housing market. • UGC prices for long-term profitability to absorb market disruptions and has cumulatively generated $2.6 billion in pre-tax operating income and maintained a cumulative loss ratio of 50% for the 10-year period ended December 31, 2007. • The quality of new business production is improving, driven by UGC’s underwriting and eligibility adjustments, along with more rigorous underwriting standards applied by UGC’s lender customers. The positive effects of these changes will be reflected in future years’ results. • UGC is well positioned to take advantage of the opportunities presented when the market emerges from this housing correction. Executive Statement
  • 95. 94 United Guaranty • Mortgage Guaranty is a multi-year contract with monthly premiums and automatic renewals (15-30 year mortgage term). UGC can generally only cancel the policy for non-payment of premium or other policy exclusions. • Mortgage Guaranty first-lien price changes are slow to affect results, as they are regulated by state Departments of Insurance and require changes to loan origination systems by large mortgage lenders. • Mortgage Guaranty performance is predominantly determined by macroeconomic events in the early years of the policy. Current year loss expenses are driven by loans from prior vintage years. • This business model results in a portfolio with an average life of 5-7 years, with new production contributing less than 20% of the calendar year net premiums written but building a base for renewal premiums. Mortgage Guaranty Product Characteristics Changes in underwriting and eligibility guidelines positively affect future results, but do not have a significant effect on current year results.
  • 96. 95 United Guaranty December 31, 2007 This table is for informational purposes only. Net Risk-in-Force (RIF) = Insurance risk on mortgages net of risk sharing and reinsurance. Loans with unknown FICO scores are included in the FICO (620-659) based on similar performance characteristics. Delinquency figures are based on number of policies (not dollar amounts), consistent with mortgage insurance industry practice. Real Estate Portfolio Total Portfolio FICO (≥ 660) FICO (620-659) FICO (<620) Domestic Mortgage Net Risk-in-Force 60+ Day Delinquency $29.8 Billion 3.71% $21.0 Billion 2.10% $6.3 Billion 6.60% $2.5 Billion 15.56% 2007 Vintage 60+ Day Delinquency $8.9 Billion 2.46% $6.2 Billion 1.07% $1.9 Billion 3.96% $886 Million 12.99% 2006 Vintage 60+ Day Delinquency $6.5 Billion 4.47% $4.5 Billion 2.65% $1.4 Billion 7.68% $670 Million 18.12% 2005 Vintage 60+ Day Delinquency $5.1 Billion 3.69% $3.7 Billion 2.39% $1.0 Billion 7.10% $292 Million 15.17% 2004 Vintage 60+ Day Delinquency $3.3 Billion 3.36% $2.4 Billion 1.73% $701 Million 6.81% $214 Million 16.80% Loans > 95% 60+ Day Delinquency $10.4 Billion 4.32% $6.6 Billion 2.13% $2.6 Billion 7.38% $1.1 Billion 16.44% Low Documentation 60+ Day Delinquency $5.6 Billion 3.89% $5.0 Billion 3.25% $532 Million 8.10% $112 Million 18.29% Interest Only & Option ARMs 60+ Day Delinquency $2.9 Billion 8.82% $2.4 Billion 7.76% $433 Million 13.07% $81 Million 16.68%
  • 97. 96 United Guaranty • The deterioration of the U.S. housing market has affected all segments of the mortgage business, but the high LTV second-lien product is particularly sensitive to declining home values and as a result constitutes a disproportionate share of 2007 losses. • Due to the accelerated claims cycle for second-lien mortgages, these net losses incurred should work through the portfolio much faster and peaked in 2007. • However, first-lien net losses incurred are beginning to negatively impact operating results as delinquencies progress through the claim cycle. Further deterioration is anticipated in 2008. • As of 12/31/07, expected future losses are significantly below net risk-in-force. Future premiums are expected to exceed future losses on the existing portfolio. Domestic First Lien - $697MM 49% of losses incurred Domestic Second Lien - $737MM 51% of losses incurred Domestic Second Lien - $3.8BN 13% of portfolio Domestic First Lien - $26.0BN 87% of portfolio United Guaranty Domestic Mortgage Net Risk-In-Force December 31, 2007 Although quality of new business production is improving, near-term results will continue to reflect market downturn. Loss Emergence Domestic Second Lien - $3.8BN 13% of portfolio Domestic First Lien - $26.0BN 87% of portfolio United Guaranty Domestic Mortgage Net Risk-In-Force December 31, 2007 Domestic First Lien - $697MM 49% of net losses incurred Domestic Second Lien - $737MM 51% of net losses incurred United Guaranty Domestic Mortgage Net Losses Incurred Total Year 2007
  • 98. 97 United Guaranty Analysis of Loss Reserve – Domestic Mortgage Product • UGC’s Corporate Actuarial Department employs rigorous analyses of the loss reserve adequacy of its businesses on a quarterly basis - The total loss reserve equals the sum of the case reserves and incurred but not reported (IBNR) reserves. • In the actuarial testing of loss reserve adequacy, a variety of data and methods are employed - Accident year data is the primary focus, which represents the date of first missed payment on a loan. - Reserving methods typically include: paid development, incurred development, Cape Cod, Bornhuetter-Ferguson and incurred count severity. - A range of reserve estimates is established based on observed historical variance in loss reserve estimates and a selected confidence level. - An updated analysis of the case reserve and IBNR factors is performed on a quarterly basis. • The actuarial analysis results, together with any recommended changes in reserves, are reviewed on a quarterly basis and approved by UGC’s CFO, Controller and Chief Risk Officer, as well as by AIG’s Chief Actuary, Chief Credit Officer and the CFO of AIG’s Domestic Brokerage Group. Mortgage Guaranty Insurance accounting requires reserves to be established based upon current delinquencies, but does not permit any provision for future delinquencies.
  • 99. 98 • $26.0 billion net risk-in-force • 878,829 policies in force • Average FICO score of 696 • 48,263 delinquent loans • 5.49% delinquency ratio United Guaranty First-Lien Portfolio In-Force Summary as of December 31, 2007
  • 100. 99 4.94 5.01 4.69 4.41 4.26 4.29 4.29 4.39 4.49 4.51 4.59 4.62 4.68 4.92 4.73 4.52 4.44 4.52 4.68 4.89 5.06 5.28 5.65 5.93 6.33 3.70 3.76 3.51 3.26 3.14 3.20 3.26 3.36 3.39 3.48 3.56 3.59 3.72 3.91 3.74 3.56 3.56 3.71 3.98 4.23 4.39 4.65 4.89 5.30 5.69 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00 Dec- 05 Jan- 06 Feb- 06 Mar- 06 Apr- 06 May- 06 Jun- 06 Jul- 06 Aug- 06 Sep- 06 Oct- 06 Nov- 06 Dec- 06 Jan- 07 Feb- 07 Mar- 07 Apr- 07 May- 07 Jun- 07 Jul- 07 Aug- 07 Sep- 07 Oct- 07 Nov- 07 Dec- 07 Industry (excluding UGC, Radian) Domestic First-Lien United Guaranty Delinquency Rates – UGC vs. Industry (First-Lien Primary) Industry* United Guaranty The first-lien mortgage delinquency ratio has consistently run below the industry average, although the gap is narrower than historical levels. *Source: Mortgage Insurance Companies of America (MICA) Figures (for UGC and industry) are based on primary insurance and do not include pool insurance. %
  • 101. 100 First-Lien Portfolio by Product 78% of the first-lien portfolio is in fixed rate loans. Portfolio Net RIF by Product as of December 31, 2007 $1.1BN 4% $20.2BN 78% $2.9BN 11% $1.8BN 7% Fixed Rate Loans Positively Amortizing ARM’s Potential Neg-Am ARM’s Interest Only Loans United Guaranty
  • 102. 101 First-Lien Portfolio by Flow / Bulk Channel UGC chose to be a minor participant in the high risk bulk channel. Portfolio Net RIF by Channel as of December 31, 2007 $1.2BN 5% $24.8BN 95% Flow Channel Bulk Channel United Guaranty
  • 103. 102 First-Lien Key Risk Initiatives UGC continues to implement key risk initiatives to improve the quality of new business production. • Tightened underwriting standards and eligibility guidelines in conjunction with market movement. • Increased rates in select, higher risk business segments. • Established new and modified existing portfolio concentration caps. • Flight to quality: – Improved mortgage insurance penetration (fewer “piggybacks”). – Increased conforming (GSE eligible) loan production. – Improved quality of new business production. United Guaranty
  • 104. 103 • $ 3.8 billion net risk-in-force • 643,147 policies in force • Average FICO score of 716 • 8,205 delinquent loans • 1.28% delinquency ratio In-Force Summary as of December 31, 2007 United Guaranty Second-Lien Portfolio
  • 105. 104 88% of the second-lien portfolio is in high credit quality loans. Portfolio Net RIF by FICO score as of December 31, 2007 $3.4BN 88% $0.4BN 11% FICO ≥ 660 FICO 620 – 659 FICO < 620 $59MM 1% United Guaranty Second-Lien Portfolio by FICO Score
  • 106. 105 • Strengthened Underwriting Guidelines – Eliminated Alternative Risk product in fourth quarter 2006. – Eliminating 100% Combined LTV (CLTV) purchase money (“piggyback”) seconds. – Eliminated significant segments of stated income and third party originated business. • Reduced Risk Retention Levels – In lieu of 100% coverage, introducing co-insurance to align the lenders’ interests with those of UGC. – Utilizing mezzanine risk layers and lower policy limits (policy limit is commonly referred to as stop loss in the mortgage insurance industry). • Improved Pricing – Implementing higher pricing on new business. – Utilizing experience and retrospective rating plans more frequently. • Enhanced Portfolio Risk Management – Established several new portfolio concentration caps in addition to modifying selected existing caps. UGC continues to implement key risk initiatives to improve the quality of new business production. United Guaranty Second-Lien Key Risk Initiatives
  • 107. 106 United Guaranty Risk Mitigating Factors • UGC uses several mitigants to reduce the volatility of losses transferred from lenders, which are reflected in the net risk-in-force figures: - Risk sharing: funded arrangements through captive reinsurance with most major lenders, in which the lenders share in losses above a determined attachment point. - Reinsurance: quota share reinsurance on a portion of UGC’s sub-prime first-lien product and segments of its second-lien product. - Policy limit: second-lien mortgage business has an aggregate policy limit provision limiting losses to a percentage (generally 10%) of the total original loan balances in each policy. - Fraud exclusion: UGC maintains a fraud exclusion on both its first-lien and its second-lien mortgage businesses. • 78% of first-lien mortgages are fixed rate, which have about 50% lower delinquency than ARMs. • First-lien mortgages consist of 87% single family residences and 93% owner occupied.
  • 108. 107 United Guaranty • UGC is engaged in a highly cyclical business, with high returns in 8 out of 10 years and underwriting losses in 2 out of 10 years on average. The downturns in the housing industry negatively affect short-term profitability, as pricing is actuarially derived for the long-term performance. • UGC expects that the downward market cycle will continue to adversely affect its operating results for the foreseeable future and is likely to result in another significant operating loss in 2008. • UGC has re-engineered its second-lien product, further tightened first-lien eligibility guidelines and increased rates in select, high-risk business segments. • The quality of new business production is improving, driven by UGC’s underwriting and eligibility adjustments, along with more rigorous underwriting standards applied by UGC’s lender customers. The positive effects of these changes will be reflected in future years’ results. • Additional positive market trends include: • Improved mortgage insurance market penetration. • Increased conforming (GSE eligible) loan products. • Improved persistency of insured portfolio. Summary
  • 110. 109 American General Finance • For over 50 years, AIG’s domestic consumer finance business has provided mortgage and consumer loans through a network of branch offices, which currently consists of 1,600 locations. • In addition, AGF originates and acquires mortgage loans through its centralized mortgage operations.
  • 111. 110 $0 $5 $10 $15 $20 $25 $30 YE04 YE05 YE06 YE07 Real Estate Non-Real Estate Retail Sales Finance $ Billions $20.2 7% 15% 6% 14% $23.8 80% 8% 14% 78% $24.3 $25.5 9% 15% 76% 78% Total Receivables Before Allowance American General Finance Portfolio Mix
  • 112. 111 American General Finance December 31, 2007 This table is for informational purposes only. AGF’s loan underwriting process does not use FICO scores as a primary determinant for credit decisions. AGF uses proprietary risk scoring models in making credit decisions. Delinquency figures are shown as a percentage of outstanding loan balances, consistent with mortgage lending practice. Differences in totals by columns and rows are due to rounding. Outstandings LTV 80% 84% 80% 75% 60+% 2.64% 1.29% 3.05% 4.48% 2007 Vintage LTV 77% 82% 79% 74% 60+% 0.99% 0.44% 1.01% 1.33% 2006 Vintage LTV 80% 87% 81% 75% 60+% 3.57% 1.70% 3.02% 5.26% 2005 Vintage LTV 82% 85% 82% 76% 60+% 3.03% 1.65% 4.49% 5.60% 2004 Vintage LTV 82% 83% 80% 75% 60+% 1.98% 1.12% 3.71% 6.06% LTV Greater than 95.5% LTV 99% 99% 99% 98% 60+% 2.36% 1.81% 5.10% 5.69% Low Documentation LTV 76% 78% 76% 71% 60+% 4.30% 3.86% 4.22% 5.97% Interest Only LTV 89% 89% 88% 79% 60+% 3.12% 2.55% 5.12% 11.90% Total Portfolio FICO (≥ 660) FICO (620 - 659) FICO (< 620) $19.5 Billion $9.7 Billion $3.3 Billion $6.2 Billion $4.0 Billion $1.3 Billion $830.7 Million $1.9 Billion $3.4 Billion $1.2 Billion $647.6 Million $1.5 Billion $4.8 Billion $2.9 Billion $856.0 Million $1.1 Billion $4.5 Billion $3.5 Billion $554.0 Million $503.1 Million $3.5 Billion $3.0 Billion $385.7 Million $169.4 Million $531.9 Million $283.5 Million $165.4 Million $83.0 Million $1.7 Billion $1.4 Billion $284.3 Million $23.8 Million
  • 113. 112 American General Finance Net Real Estate Loan Growth $ Billions $1.4 $1.2 $0.4 $0.3 $0.1 $0.1 -$0.1 -$0.3 $0.0 $0.2 $0.3 $0.0 -$0.6 $0.0 $0.6 $1.2 $1.8 1Q05 2Q05 3Q05 4Q05 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 As the real estate market softened, AGF maintained its underwriting discipline despite experiencing lower volume and growth.
  • 114. 113 0% 5% 10% 15% 20% 25% 2004 2005 2006 1Q07 2Q07 3Q07 4Q07 AGF Total Real Estate 60+ Day Delinquency*60+ Day Delinquency* * Source: Company and industry reports Subprime ABS Real Estate MarketSubprime ABS Real Estate Market American General Finance AGF vs. “Subprime ABS Market”
  • 115. 114 0% 1% 2% 3% 4% 5% YE03 YE04 YE05 YE06 YE07 American General Finance Real Estate Credit Quality AGF’s delinquency and losses have risen from recent all- time lows, yet remain below their targets as well as the industry during the current mortgage market conditions. Net Charge-off Target .75% - 1.25% 60+ Day Delinquency Target 3.0% - 4.0% Target ranges were set by AGF management years ago to denote sound credit quality parameters.
  • 116. 115 American General Finance Risk Mitigating Practices - Real Estate Portfolio • 97% of mortgages are underwritten with full income verification. • 88% are fixed-rate mortgages; only about 9% of the total mortgage portfolio re-sets interest rates by the end of 2008; about 10% by the end of 2009. • Adjustable rate mortgages (ARMs): borrowers are qualified on a fully-indexed and fully-amortizing basis at origination. • No delegation of underwriting to unrelated parties. • No Option ARMs. • Substantially all loans are: – First mortgages (92%) – Owner occupant borrowers (94%) • Geographically diverse portfolio.
  • 117. 116 • AGF’s allowance for loan losses is maintained at a level considered adequate to absorb management’s best estimate of credit losses in the existing portfolio. • AGF’s Credit Strategy and Policy Committee is responsible for determining the appropriate level for the allowance. – Membership consists of AGF’s senior management, including, among others, AGF’s CEO, the Executive Vice President of AGF’s Branch Operations, AGF’s CFO and AGF’s Chief Risk Officer – The Committee evaluates both internal and external factors including: • The composition of AGF’s finance receivable portfolio • Prior finance receivables losses and delinquency experience • Results of migration and Monte Carlo analyses • Current economic environment • AGF calculates three different migration scenarios based on varying assumptions to evaluate a range of possible outcomes for the quantitative component of the allowance for residential real estate. • Conclusions reached by the Committee are reviewed on a quarterly basis and approved by AIG’s Chief Credit Officer and the CFO of AIG’s Financial Services Division. American General Finance Allowance Methodology
  • 118. 117 0% 1% 2% 3% 4% 5% 2004 2005 2006 2007 0% 1% 2% 3% 4% 5% Charge-Offs Allowance for Loan Loss Total PortfolioTotal Portfolio 1.6X 2.0X 2.1X Coverage Ratio ($ Allowance / $ Charge-offs) 2.1X American General Finance Strong Credit Quality
  • 119. 118 American General Finance Equity One Portfolio Acquisition • AGF has agreed to acquire $1.5 billion in outstanding balances of branch- based consumer loans of Equity One, Incorporated. • Equity One’s branch-offered products are similar to those of AGF (1st & 2nd Fixed Rate Mortgages, Consumer Loans, Retail Sales). • Customer profile and credit quality performance are also very similar to those of AGF. • The transaction is expected to close during the first quarter of 2008. • Acquisition of 130,000 new accounts with minimal overhead increase. • AGF was able to acquire the portfolio at an attractive price given current market conditions. • Acquisition is expected to be accretive to earnings in 2008.
  • 120. 119 American General Finance Summary • At the end of the fourth quarter, AGF’s real estate loan portfolio remained at $19.5 billion, flat to the end of the third quarter. – The 2007 vintage production is the result of balanced growth from both its centralized real estate and branch operations which met both strict underwriting guidelines and return hurdles in a challenging market. • AGF maintained its time-tested, disciplined underwriting approach throughout the residential real estate boom, continually re-evaluating guidelines and adjusting as appropriate, resulting in: – Lower volume – Better than targeted delinquency and charge-off – Better than industry-experienced delinquency and charge-off • AGF believes that the housing market will likely continue to deteriorate for the remainder of 2008, but the company’s business model and strict underwriting approach are sound, allowing the company to pursue opportunities.
  • 123. A-2 Summary StatisticsSummary Statistics ““Super SeniorSuper Senior”” Credit Derivatives*Credit Derivatives* Transaction Type Corporate – Regulatory Capital Motivated European Mortgage Corporate - Arbitrage Motivated Multi-Sector CDOs Transactions w/Mixed Transactions Collateral, including w/No Sub Prime Sub Prime Collateral Gross Notional ($ Billion) 306.0 182.8 87.3 82.8 27.3 AIGFP Net Notional Exposure ($ Billion) 229.6 149.1 70.4 61.4 16.8 Number of Transactions 58 35 36 103 13 Weighted Average Subordination (%) ¹ 22.0% 13.8% 18.3% 23.3% 18.0% Average Number of Obligors / Transaction 1,571 74,819 122 194 185 Expected Maturity (Years) 1.2 ² 2.3 ² 4.0 5.0 ³ 5.4 ³ 1. Weighted by Gross Transaction Notional 2. Maturity shown reflects first non regulatory call date, although majority of transactions have Regulatory Capital Calls from Jan 08 3. Reflects the Weighted Average Life *All data is as of December 31, 2007.
  • 124. A-3 Summary of Key Statistics for Multi SectorSummary of Key Statistics for Multi Sector CDOsCDOs withwith Mixed Collateral, Including Sub PrimeMixed Collateral, Including Sub Prime (Billion US$)(Billion US$) • Total Gross Transaction Notional: 53.1 • AIGFP Total Net* Notional: 43.5 • Number of Transactions: 45 • Average Attachment: 15.4% Sub Prime Reference Obligations • Gross Sub Prime Notional: 25.2 • AIGFP Net* Sub Prime Notional: 17.3 • Average Sub Prime: 47.5% • Avg. HPA for Sub Prime Collateral: 7.7% • Average Cum. Loss Rate: 1.2% • Average FICO: 625 • Total Gross Transaction Notional: 29.7 • AIGFP Total Net* Notional: 17.9 • Number of Transactions: 58 • Average Attachment: 37.3% Sub Prime Reference Obligations • Gross Sub Prime Notional: 18.9 • AIGFP Net* Sub Prime Notional: 8.2 • Average Sub Prime: 63.6% • Avg. HPA for Sub Prime Collateral: 7.6% • Average Cum. Loss Rate: 1.1% • Average FICO: 624 High Grade Collateral (predominantly AA Rated) Mezzanine Collateral (predominantly BBB Rated) * Net of all Transaction subordination Information shown is sourced from LoanPerformance except in the following circumstances: • Underlying Reference Obligation data is sourced from Intex • Loss Data is sourced from CSFB’s “Locus” system • Sector Categorization is compiled from Individual Transaction Trustee Reports and Moody’s and S&P data
  • 125. A-4 ““Super SeniorSuper Senior”” ExposureExposure Corporate LoansCorporate Loans
  • 126. A-5 Regulatory CapitalRegulatory Capital -- Corporate by Primary JurisdictionCorporate by Primary Jurisdiction Net Notional (Bn) Percentage of Total Current Average Subordination Realised Pool Losses to Date % Weighted Average Maturity (years) Number of Deals Primarily Single Country Exposure Portfolio To First Call * To Maturity Germany 19.9 8.7 24.7 0.09 2.2 8.9 10 USA 7.1 3.1 40.0 0.00 0.8 12.3 1 Netherlands 4.7 2.0 18.2 0.00 2.0 46.0 1 Portugal 4.2 1.8 11.6 0.08 0.8 11.8 1 UK 2.2 1.0 25.0 0.00 1.0 13.8 1 France 2.1 0.9 21.4 0.00 1.0 1.0 1 Australia/New Zealand 1.8 0.8 9.0 0.00 1.7 3.2 1 Finland 1.2 0.5 18.0 0.00 1.0 7.0 1 Belgium 3.7 1.6 28.4 0.05 1.5 6.1 2 46.9 Regional Exposure Portfolio USA Majority 66.5 29.0 25.6 0.00 1.4 5.9 15 W. Europe Majority 104.5 45.5 18.3 0.00 0.9 8.6 16 Asia/Australia Majority 10.7 4.6 16.8 0.00 0.7 3.1 6 Emerging Market 1.0 0.5 29.0 0.00 3.3 3.3 2 182.7 Total 229.6 100.0 22.0 0.01 1.2 8.4 58 * The vast majority of deals have regulatory calls from January 2008. We expect that these calls will be exercised over the next 12-18 months as the different originating banks in Europe are able to adopt the new Basle II Capital standards. The call date listed in the chart is the first non regulatory call.
  • 127. A-6 Arbitrage MotivatedArbitrage Motivated -- Corporate by Primary JurisdictionCorporate by Primary Jurisdiction Net Notional (Bn) Percentage of Total Current Average Subordination Realised Pool Losses to Date % Weighted Average Maturity (years) Number of Deals Primarily Single Country Exposure Portfolio To First Call To Maturity USA 50.2 71.3 16.8 0.08 4.7 4.7 23 Regional Exposure Portfolio USA Majority 18.1 25.7 15.7 0.69 1.7 1.7 7 W. Europe Majority 2.1 3.0 49.2 0.00 5.0 14.1 6 20.2 Total 70.4 100.0 18.3 0.22 4.0 4.5 36
  • 128. A-7 ““Super SeniorSuper Senior”” ExposureExposure European Residential MortgagesEuropean Residential Mortgages
  • 129. A-8 European Residential Mortgages Summary by GeographyEuropean Residential Mortgages Summary by Geography AIGFP Net Notional Exposure (Bn) % of Total Exposure Current Average Subordination % Realised Losses to Date % of Pool Weighted Average Maturity (years) Number of Transactions To First Call* To Maturity Denmark 38.9 26.2 9.2 0.00 1.4 31.8 3 France 38.0 25.5 8.2 0.01 2.0 31.6 7 Germany 36.3 24.3 19.0 0.10 1.7 42.1 17 Netherlands 22.8 15.3 21.7 0.01 3.8 10.2 4 Sweden 9.6 6.4 12.9 0.00 2.0 34.5 2 UK 1.8 1.2 10.0 0.00 1.2 31.2 1 Spain 1.7 1.1 9.2 0.00 9.5 42.1 1 Total 149.1 100.0 13.8 0.03 2.3 30.3 35 * All of these deals have regulatory calls from January 2008. We expect that these calls will be exercised over the next 12-18 months as the different originating banks in Europe are able to adopt the new Basle II Capital standards. The call date listed in the chart is the first non regulatory call.
  • 130. A-9 ““Super SeniorSuper Senior”” Multi Sector CDO ExposureMulti Sector CDO Exposure Consisting of Mixed Collateral,Consisting of Mixed Collateral, including Sub Prime:including Sub Prime: Mezzanine Collateral Underlying SummaryMezzanine Collateral Underlying Summary
  • 131. A-10 Summary of Transaction Current StatusSummary of Transaction Current Status Gross Notional 29.7bn AIG Notional Exposure (net of transaction subordination) 17.9bn Max: 904 mm Min: 31 mm Avg: 309 mm Number of Transactions 58 Managed Transactions 21 Static Transactions 37 Average Number of Obligors 172 Average AIG Attachment Point (weighted by transaction notional) 37.3% Max: 74.8% Min: 24.5% Average % of AIG Subordination that is AAA Rated by at least one agency 36.7% Max: 62.3% Min: 7.4% Average Subordinated AAA Tranche Thickness 13.9% Max: 31.2% Min: 2.4% Number of Transactions that are Amortizing 43 Information used in this presentation is sourced from LoanPerformance except in the following circumstances: • Underlying Reference Obligation data is sourced from Intex • Rating Information is sourced from Bloomberg • Loss Data is sourced from CSFB’s “Locus” system • Sector Categorization is compiled from Individual Transaction Trustee Reports and Moody’s and S&P data “Super Senior” Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Data as of Dec 31
  • 132. A-11 Summary of Reference Obligations in ourSummary of Reference Obligations in our Transactions:Transactions: Sub Prime Other RMBS CDO CMBS Other ABS Non ABS Total Classifications 63.6% 17.4% 6.4% 7.5% 4.2% 0.9% 100.0% AAA AA A BBB BB <BB NR Rating 2.4% 4.7% 11.6% 55.2% 11.5% 13.6% 1.0% 100.0% Percentages shown are of Gross Transaction Notional “Super Senior” Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary
  • 133. A-12 Recent Rating PerformanceRecent Rating Performance Underlying Reference Obligations in our Transactions Moody’s S&P % of Sub Prime Upgraded Since Deal Inception 1.2% 0.6% % of Sub Prime Downgraded Since Deal Inception 17.1% 21.0% Overall Transaction Deals with Junior Tranches on Negative Review 27 33 Junior Tranches Downgraded 10 25 AIG Tranche Downgraded None None “Super Senior” Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary
  • 134. A-13 Underlying Sub Prime Reference ObligationsUnderlying Sub Prime Reference Obligations AAA AA A BBB BB <BB Total Rating 0.5% 1.4% 4.9% 36.4% 9.4% 11.0% 63.6% Pre 2004 2004 2005 2006 2007 Total Vintage 3.3% 17.8% 34.6% 4.4% 3.5% 63.6% Q1 2005 Q2 2005 Q3 2005 Q4 2005 Q1 2006 Q2 2006 Q3 2006 Q4 2006 Q1 2007 Q2 2007 Q3 2007 Q4 2007 Total Vintage by Quarter 13.8% 11.6% 5.9% 3.3% 1.5% 1.3% 0.7% 0.9% 1.4% 1.3% 0.5% 0.3% 42.5% AAA AA A BBB BB <BB Total 2006 Vintage 0.1% 0.4% 0.2% 0.9% 0.6% 2.2% 4.4% AAA AA A BBB BB <BB Total 2007 Vintage 0.2% 0.4% 0.4% 0.7% 0.8% 1.0% 3.5% “Super Senior” Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
  • 135. A-14 Additional Sub Prime BreakdownAdditional Sub Prime Breakdown Avg. % of Sub Prime Collateral that is 2nd Lien 3.5% Avg. % of Sub Prime Collateral with 2nd Lien >90% 0.8% Average LTV at Inception 80.6 Avg. HPA on Reference Sub Prime Bonds 7.6% Max: 22.8% Min: 3.6% Current Average 12 mos CPR Rate 29.8% Average FICO Score 624 Floating 2 Yr ARM 3 Yr ARM Fixed Average Loan Type 61.3% 5.8% 6.7% 38.7% Sub Prime Originators New Century 5.4% State Concentration California 22.5% Ameriquest 4.8% Florida 7.2% Countrywide 3.7% Texas 6.1% Option 1 3.7% New York 5.8% Fremont 2.9% Michigan 2.0% Current Weighted Average Loss Rate on Sub Prime 1.1% Percentages shown are of Total Sub Prime “Super Senior” Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary
  • 136. A-15 Underlying CDO Reference Obligations BreakdownUnderlying CDO Reference Obligations Breakdown High Grade Mezz Other ABS CMBS Other Total Classifications 0.7% 3.3% 0.3% 0.8% 1.3% 6.4% AAA AA A BBB BB <BB NR Total Rating 0.2% 0.5% 1.1% 2.5% 0.6% 1.3% 0.2% 6.4% Pre 2003 2003 2004 2005 2006 2007 Total Vintage 0.3% 0.5% 2.0% 2.6% 0.8% 0.2% 6.4% Other RMBS by Vintage Pre 2003 2003 2004 2005 2006 2007 Total 0.2% 2.0% 6.2% 7.2% 1.4% 0.4% 17.4% “Super Senior” Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
  • 137. A-16 ““Super SeniorSuper Senior”” Multi Sector CDO ExposureMulti Sector CDO Exposure Consisting of Mixed Collateral,Consisting of Mixed Collateral, including Sub Prime:including Sub Prime: High Grade Collateral Underlying SummaryHigh Grade Collateral Underlying Summary
  • 138. A-17 Summary of Transaction Current StatusSummary of Transaction Current Status Gross Notional 53.1bn AIG Notional Exposure (net of transaction subordination) 43.5bn Max: 2.2 bn Min: 44.8 mm Avg: 964 mm Number of Transactions 45 Managed Transactions 20 Static Transactions 25 Average Number of Obligors 206 Average AIG Attachment Point 15.4% Max: 46.2% Min: 10.0% Average % of AIG Subordination that is AAA Rated 43.0% Max: 86.1% Min: 0.0% Average Subordinated AAA Tranche Thickness 7.4% Max: 39.7% Min: 0.0% Number of Transactions that are Amortizing 32 “Super Senior” Multi Sector CDO Exposure – High Grade Collateral Underlying SummaryData as of Dec 31 Information used in this presentation is sourced from LoanPerformance except in the following circumstances: • Underlying Reference Obligation data is sourced from Intex • Rating Information is sourced from Bloomberg • Loss Data is sourced from CSFB’s “Locus” system • Sector Categorization is compiled from Individual Transaction Trustee Reports and Moody’s and S&P data
  • 139. A-18 Summary of Underlying Reference Obligations in ourSummary of Underlying Reference Obligations in our Transactions:Transactions: Sub Prime Other RMBS CDO CMBS Other ABS Non ABS Total Classifications 47.5% 27.9% 15.2% 6.8% 2.4% 0.2% 100.0% AAA AA A BBB BB <BB NR Rating 28.2% 35.7% 26.1% 4.1% 2.4% 1.1% 2.4% 100.0% “Super Senior” Multi Sector CDO Exposure – High Grade Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
  • 140. A-19 Recent Rating PerformanceRecent Rating Performance Underlying Reference Obligations in our Transactions Moody’s S&P % of Sub Prime Upgraded Since Deal Inception 1.1% 0.9% % of Sub Prime Downgraded Since Deal Inception 1.6% 2.3% Overall Transaction Deals with Junior Tranches on Negative Review 9 14 Junior Tranches Downgraded 3 None AIG Tranche Downgraded None None “Super Senior” Multi Sector CDO Exposure – High Grade Collateral Underlying Summary
  • 141. A-20 Underlying Sub Prime Reference ObligationsUnderlying Sub Prime Reference Obligations AAA AA A BBB BB <BB Total Rating 4.6% 21.4% 18.1% 2.3% 0.8% 0.3% 47.5% Pre 2004 2004 2005 2006 2007 Total Vintage 2.2% 13.7% 26.4% 2.6% 2.6% 47.5% Q1 2005 Q2 2005 Q3 2005 Q4 2005 Q1 2006 Q2 2006 Q3 2006 Q4 2006 Q1 2007 Q2 2007 Q3 2007 Q4 2007 Total Vintage by Quarter 6.6% 9.5% 5.8% 4.5% 1.3% 0.6% 0.3% 0.4% 0.7% 0.8% 0.7% 0.4% 31.6% AAA AA A BBB BB <BB Total 2006 Vintage 0.9% 0.9% 0.5% 0.1% 0.1% 0.1% 2.6% AAA AA A BBB BB <BB Total 2007 Vintage 0.9% 0.9% 0.3% 0.4% 0.1% 0.0% 2.6% Percentages shown are of Gross Transaction Notional “Super Senior” Multi Sector CDO Exposure – High Grade Collateral Underlying Summary
  • 142. A-21 Additional Sub Prime BreakdownAdditional Sub Prime Breakdown Avg. % of Sub Prime Collateral that is 2nd Lien 3.2% Avg. % of Sub Prime Collateral with 2nd Lien >90% 0.7% Average LTV at Inception 80.6 Avg. HPA on Reference Sub Prime Bonds 7.7% Max: 22.3% Min: 3.1% Current Average 12 month CPR Rate 30.2% Average FICO Score 625 Floating 2 Yr ARM 3 Yr ARM Fixed Average Loan Type 60.2% 5.3% 7.5% 39.8% Sub Prime Originators Countrywide 4.8% State Concentration California 22.9% Ameriquest 4.5% Florida 7.4% New Century 4.0% Texas 5.7% Lehman 2.7% New York 5.7% Fremont 2.7% Michigan 2.0% Current Weighted Average Loss Rate on Sub Prime 1.2% Percentages shown are of Total Sub Prime “Super Senior” Multi Sector CDO Exposure – High Grade Collateral Underlying Summary
  • 143. A-22 Underlying CDO Reference Obligations BreakdownUnderlying CDO Reference Obligations Breakdown High Grade Mezzanine Other ABS CMBS Other Total Classifications 2.8% 8.0% 0.7% 1.1% 2.6% 15.2% AAA AA A BBB BB <BB NR Total Rating 4.0% 5.1% 2.0% 1.2% 1.4% 0.6% 0.9% 15.2% Pre 2003 2003 2004 2005 2006 2007 Total Vintage 0.8% 1.3% 6.4% 4.8% 1.5% 0.4% 15.2% Other RMBS by Vintage Pre 2003 2003 2004 2005 2006 2007 Total 0.4% 1.0% 7.1% 16.1% 2.3% 1.0% 27.9% Percentages shown are of Gross Transaction Notional “Super Senior” Multi Sector CDO Exposure – High Grade Collateral Underlying Summary
  • 144. A-23 Cash Multi Sector CDO ExposureCash Multi Sector CDO Exposure Consisting of Mixed Collateral,Consisting of Mixed Collateral, including Sub Prime:including Sub Prime: Mezzanine Collateral Underlying SummaryMezzanine Collateral Underlying Summary
  • 145. A-24 Summary of Transaction Current StatusSummary of Transaction Current Status AIG Notional Exposure (net of transaction subordination) 2.6bn Max: 191 mm Min: 4 mm Avg: 41 mm Number of Transactions 63 Managed Transactions 21 Static Transactions 42 Average Number of Obligors 162 Average AIG Attachment Point (weighted by transaction notional) 28.5% Max: 59.6% Min: 12.0% Number of Transactions that are Amortizing 34 Rating of Exposure Moody’s S&P AAA 47 58 AA 2 5 NR 14 0 Information used in this presentation is sourced from LoanPerformance except in the following circumstances: • Underlying Collateral Data is sourced from Intex • Rating Information is sourced from Bloomberg • Loss Data is sourced from CSFB’s “Locus” system • Sector Categorization is sourced from Individual Transaction Trustee Reports along with Moody’s and S&P data Cash Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary
  • 146. A-25 Summary of Reference Obligations in ourSummary of Reference Obligations in our Transactions:Transactions: Sub Prime Other RMBS CDO CMBS Other ABS Non ABS Total Classifications 65.8% 16.7% 4.7% 7.2% 5.0% 0.6% 100.0% AAA AA A BBB BB <BB NR Rating 2.6% 5.7% 13.7% 51.4% 12.4% 13.4% 0.8% 100.0% Cash Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
  • 147. A-26 Recent Rating PerformanceRecent Rating Performance Underlying Reference Obligations in our Transactions Moody’s S&P % of Sub Prime Upgraded Since Deal Inception 2.2% 1.7% % of Sub Prime Downgraded Since Deal Inception 29.8% 36.2% Overall Transaction Deals with Junior Tranches on Negative Review 18 18 Junior Tranches Downgraded 6 13 AIG Tranche Downgraded None 5 Cash Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary
  • 148. A-27 Underlying Sub Prime Reference ObligationsUnderlying Sub Prime Reference Obligations AAA AA A BBB BB <BB Total Rating 0.4% 1.2% 6.7% 35.6% 10.5% 11.4% 65.8% Pre 2004 2004 2005 2006 2007 Total Vintage 3.6% 23.1% 32.8% 3.5% 2.8% 65.8% Q1 2005 Q2 2005 Q3 2005 Q4 2005 Q1 2006 Q2 2006 Q3 2006 Q4 2006 Q1 2007 Q2 2007 Q3 2007 Q4 2007 Total Vintage by Quarter 14.3% 10.7% 5.1% 2.7% 1.2% 0.9% 0.5% 0.9% 1.1% 0.9% 0.5% 0.3% 39.1% AAA AA A BBB BB <BB Total 2006 Vintage 0.1% 0.3% 0.1% 0.7% 0.5% 1.8% 3.5% AAA AA A BBB BB <BB Total 2007 Vintage 0.1% 0.3% 0.3% 0.8% 0.6% 0.7% 2.8% Cash Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
  • 149. A-28 Additional Sub Prime BreakdownAdditional Sub Prime Breakdown Avg. % of Sub Prime Reference Collateral that is 2nd Lien 3.6% Avg. % of Sub Prime Reference Collateral with 2nd Lien >90% 0.9% Average LTV at Inception 80.92 Avg. HPA on Underlying Sub Prime Bonds 8.1% Max: 18.5% Min: 4.0% Current Average 12 mos CPR Rate 30.2% Average FICO Score 626 Floating 2 Yr ARM 3 Yr ARM Fixed Average Loan Type 59.7% 6.3% 6.9% 40.3% Top 5 Sub Prime Originators New Century 6.2% Concentration by State California 22.5% Ameriquest 5.7% Florida 6.8% Countrywide 4.2% Texas 6.3% Fremont 3.4% New York 6.0% Option One 3.4% Michigan 2.1% Current Weighted Average Loss Rate on Sub Prime 1.1% Cash Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Percentages shown are of Total Sub Prime
  • 150. A-29 Underlying CDO Reference Obligations BreakdownUnderlying CDO Reference Obligations Breakdown High Grade Mezz Other ABS CMBS Other Total Classifications 0.4% 2.4% 0.5% 0.7% 0.7% 4.7% AAA AA A BBB BB <BB NR Total Rating 0.1% 0.5% 0.6% 2.0% 0.4% 1.0% 0.1% 4.7% Pre 2003 2003 2004 2005 2006 2007 Total Vintage 0.2% 0.2% 1.0% 2.4% 0.8% 0.1% 4.7% Other RMBS by Vintage Pre 2003 2003 2004 2005 2006 2007 Total 0.3% 2.5% 7.1% 5.4% 1.2% 0.2% 16.7% Cash Multi Sector CDO Exposure - Mezzanine Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
  • 151. A-30 Cash Multi Sector CDO ExposureCash Multi Sector CDO Exposure Consisting of Mixed Collateral,Consisting of Mixed Collateral, including Sub Prime:including Sub Prime: High Grade Collateral Underlying SummaryHigh Grade Collateral Underlying Summary
  • 152. A-31 Summary of Transaction Current StatusSummary of Transaction Current Status AIG Notional Exposure (net of transaction subordination) 1.5bn Max: 250 mm Min: 14 mm Avg: 134 mm Number of Transactions 11 Managed Transactions 4 Static Transactions 7 Average Number of Obligors 202 Average AIG Attachment Point 15.0% Max: 37.0% Min: 11.4% Number of Transactions that are Amortizing 5 Rating of Exposure Moody’s S&P AAA 11 11 Information used in this presentation is sourced from LoanPerformance except in the following circumstances: • Underlying Collateral Data is sourced from Intex • Rating Information is sourced from Bloomberg • Loss Data is sourced from CSFB’s “Locus” system • Sector Categorization is sourced from Individual Transaction Trustee Reports along with Moody’s and S&P data Cash Multi Sector CDO Exposure – High Grade Collateral Underlying Summary
  • 153. A-32 Summary of Underlying Reference Obligations in ourSummary of Underlying Reference Obligations in our Transactions:Transactions: Sub Prime Other RMBS CDO CMBS Other ABS Non ABS Total Classifications 39.0% 24.4% 12.6% 20.5% 2.5% 1.0% 100.0% AAA AA A BBB BB <BB NR Rating 37.1 23.5% 22.6% 6.1% 1.9% 1.1% 7.7% 100.0% Cash Multi Sector CDO Exposure – High Grade Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
  • 154. A-33 Recent Rating PerformanceRecent Rating Performance Underlying Reference Obligations in our Transactions Moody’s S&P % of Sub Prime Upgraded Since Deal Inception 3.8% 6.1% % of Sub Prime Downgraded Since Deal Inception 4.6% 7.4% Overall Transaction Deals with Junior Tranches on Negative Review 2 None Junior Tranches Downgraded None None AIG Tranche Downgraded None None Cash Multi Sector CDO Exposure – High Grade Collateral Underlying Summary
  • 155. A-34 Underlying Sub Prime Reference ObligationsUnderlying Sub Prime Reference Obligations AAA AA A BBB BB <BB Total Rating 5.3% 15.1% 14.3% 3.4% 0.8% 0.1% 39.0% Pre 2004 2004 2005 2006 2007 Total Vintage 2.8% 9.9% 22.4% 2.9% 1.0% 39.0% Q1 2005 Q2 2005 Q3 2005 Q4 2005 Q1 2006 Q2 2006 Q3 2006 Q4 2006 Q1 2007 Q2 2007 Q3 2007 Q4 2007 Total Vintage by Quarter 3.8% 4.0% 6.2% 8.4% 1.3% 0.5% 0.6% 0.5% 0.2% 0.4% 0.3% 0.1% 26.3% AAA AA A BBB BB <BB Total 2006 Vintage 1.4% 0.6% 0.6% 0.3% 0.0% 0.0% 2.9% AAA AA A BBB BB <BB Total 2007 Vintage 0.3% 0.4% 0.3% 0.0% 0.0% 0.0% 1.0% Cash Multi Sector CDO Exposure – High Grade Collateral Underlying Summary Percentages shown are of Gross Transaction Notional
  • 156. A-35 Additional Sub Prime BreakdownAdditional Sub Prime Breakdown Avg. % of Reference Transaction Collateral that is 2nd Lien 2.4% Avg. % of Reference Transaction with 2nd Lien >90% 0.4% Average LTV at Inception 80.65 Avg. HPA on Underlying Sub Prime Bonds 6.7% Max: 16.6% Min: 3.1% Current Average 12 month CPR Rate 30.8% Average FICO Score 626 Floating 2 Yr ARM 3 Yr ARM Fixed Average Loan Type 61.5% 4.4% 7.9% 38.5% Top 5 Sub Prime Originators Countrywide 8.3% State Concentration California 24.3% Ameriquest 5.0% Florida 7.7% Lehman 5.1% Texas 5.4% New Century 4.3% New York 5.4% First Franklin 3.3% Michigan 1.7% Current Weighted Average Loss Rate on Sub Prime 1.0% Cash Multi Sector CDO Exposure – High Grade Collateral Underlying Summary Percentages shown are of Total Sub Prime
  • 157. A-36 Underlying CDO Reference Obligations BreakdownUnderlying CDO Reference Obligations Breakdown High Grade Mezzanine Other ABS CMBS Other Total Classifications 2.0% 4.8% 1.1% 2.2% 2.5% 12.6% AAA AA A BBB BB <BB NR Total Rating 5.0% 2.2% 1.8% 1.3% 0.9% 0.5% 0.9% 12.6% Pre 2003 2003 2004 2005 2006 2007 Total Vintage 1.0% 2.4% 4.0% 2.5% 2.0% 0.7% 12.6% Other RMBS by Vintage Pre 2003 2003 2004 2005 2006 2007 Total 2.2% 2.1% 6.1% 11.1% 2.2% 0.7% 24.4% Cash Multi Sector CDO Exposure – High Grade Collateral Underlying Summary Percentages shown are of Gross Transaction Notional