AIG 2009 Annual Report Download - page 161

American International Group, Inc., and Subsidiaries
actual results in any period are likely to vary, perhaps materially, from the modeled scenarios, and there can be no
assurance that the unrealized market valuation loss related to the AIGFP super senior credit default swap portfolio
will be consistent with any of the sensitivity analyses. On average, prices for CDOs declined 9.96 percent, 5.57 percent
and 0.20 percent of the notional amount outstanding for the first, second and fourth quarters of 2009, respectively.
However, for the three-month period ended September 30, 2009, prices for CDOs increased by an average of
5.13 percent of the notional amount outstanding. Further, it is difficult to extrapolate future experience based on
current market conditions.
For the purposes of estimating sensitivities for the super senior multi-sector CDO credit default swap portfolio, the
change in valuation derived using the BET model is used to estimate the change in the fair value of the derivative
liability. Out of the total $7.9 billion net notional amount of CDS written on multi-sector CDOs outstanding at
December 31, 2009, a BET value is available for $4.8 billion net notional amount. No BET value is determined for
$3.1 billion of CDS written on European multi-sector CDOs as prices on the underlying securities held by the CDOs
are not provided by collateral managers; instead these CDS are valued using counterparty prices. Therefore,
sensitivities disclosed below apply only to the net notional amount of $4.8 billion.
As mentioned above, the most significant assumption used in the BET model is the estimated price of the securities
within the CDO collateral pools. If the actual price of the securities within the collateral pools differs from the price
used in estimating the fair value of the super senior credit default swap portfolio, there is potential for material
variation in the fair value estimate. Any further declines in the value of the underlying collateral securities held by a
CDO will similarly affect the value of the super senior CDO securities given their significantly depressed valuations.
While the models attempt to predict changes in the prices of underlying collateral securities held within a CDO, the
changes are subject to actual market conditions which have proved to be highly volatile, especially given current
market conditions. AIG cannot predict reasonably likely changes in the prices of the underlying collateral securities
held within a CDO at this time.
The following table presents key inputs used in the BET model, and the potential increase (decrease) to the fair value of
the derivative liability by ABS category at December 31, 2009 corresponding to changes in these key inputs:
Average Increase (Decrease) to Fair Value of Derivative Liability
Inputs Used
(dollars in at December 31, Entire RMBS RMBS RMBS
millions) 2009 Change Portfolio PRIME ALT-A Subprime CMBS CDOs Other
Bond prices 41 points Increase of 5 points $ (314) $ (13) $ (29) $ (146) $ (81) $ (28) $ (17)
Decrease of 5 points 270 14 28 108 81 17 22
Weighted Increase of 1 year 61 2 7 52 (2) 1 1
average life 5.51 years Decrease of 1 year (121) (3) (5) (114) 4 (1) (2)
Recovery rates 23% Increase of 10% (40) (1) (1) (10) (20) (3) (5)
Decrease of 10% 50 - 2 19 26 1 2
Diversity score(a) 13 Increase of 5 (9)
Decrease of 5 34
Discount curve(b) N/A Increase of 100 bps 16
(a) The diversity score is an input at the CDO level. A calculation of sensitivity to this input by type of security is not possible.
(b) The discount curve is an input at the CDO level. A calculation of sensitivity to this input by type of security is not possible. Furthermore, for this input
it is not possible to disclose a weighted average input as a discount curve consists of a series of data points.
These results are calculated by stressing a particular assumption independently of changes in any other assumption.
No assurance can be given that the actual levels of the key inputs will not exceed, perhaps significantly, the ranges
assumed by AIG for purposes of the above analysis. No assumption should be made that results calculated from the
use of other changes in these key inputs can be interpolated or extrapolated from the results set forth above.
153 AIG 2009 Form 10-K