Freddie Mac 2010 Annual Report Download - page 233

guarantee fees, with market input assumptions extracted from the dealer quotes provided on the more liquid products,
reduced by an estimated liquidity discount.
The fair values at the time of securitization and subsequent fair value measurements at the end of a period were
primarily estimated using third-party information. Consequently, we derived the assumptions presented in Table 11.2 by
determining those implied by our valuation estimates, with the IRRs adjusted where necessary to align our internal models
with estimated fair values determined using third-party information. However, prepayment rates are presented based on our
internal models and were not similarly adjusted. For the portion of our guarantee asset that was valued by obtaining dealer
quotes on proxy securities, we derived the assumptions from the prices we were provided. Table 11.2 contains estimates of
the key assumptions used to derive the fair value measurement that related solely to our guarantee asset on financial
guarantees of single-family loans. These represent the average assumptions used both at the end of the period as well as the
valuation assumptions at guarantee issuance during the year presented on a combined basis.
Table 11.2 — Key Assumptions Used in Measuring the Fair Value of Guarantee Asset
(1)
Mean Valuation Assumptions 2009 2008
For the Year Ended
December 31,
IRRs
(2)
.................................................................................. 13.8% 12.3%
Prepayment rates
(3)
.......................................................................... 26.4% 15.5%
Weighted average lives (years) .................................................................. 3.3 5.6
(1) Estimates based solely on valuations on our guarantee asset associated with single-family loans, which represent approximately 97% of the total
guarantee asset.
(2) IRR assumptions represent a UPB weighted average of the discount rates inherent in the fair value of the recognized guarantee asset. We estimated the
IRRs using a model which employs multiple interest rate scenarios versus a single assumption.
(3) Although prepayment rates are simulated monthly, the assumptions above represent annualized prepayment rates based on UPB.
The objective of the sensitivity analysis below is to present our estimate of the financial impact of an unfavorable
change in the input values associated with the determination of fair values of these retained interests. We did not use these
inputs in determining fair value of our retained interests as our measurements were principally based on third-party pricing
information. The weighted average assumptions within Table 11.3 represent our estimates of the assumed IRR and
prepayment rates implied by market pricing as of each period end and were derived using our internal models. Since we did
not use these internal models for determining fair value in our reported results under GAAP, this sensitivity analysis is
hypothetical and would not be indicative of actual results, particularly due to the change in accounting standards on
January 1, 2010. In addition, the effect of a variation in a particular assumption on the fair value of the retained interest was
estimated independently of changes in any other assumptions. Changes in one factor would have resulted in changes in
another, which would have affected the impact of the change.
Table 11.3 — Sensitivity Analysis of Retained Interests
As of December 31, 2009
(dollars in millions)Retained Interests, Mortgage-Related Securities
Weighted average IRR assumptions ............................................................ 4.5%
Impact on fair value of 100 bps unfavorable change ............................................... $(3,634)
Impact on fair value of 200 bps unfavorable change ............................................... $(7,008)
Weighted average prepayment rate assumptions ................................................... 11.4%
Impact on fair value of 10% unfavorable change . . ............................................... $ (85)
Impact on fair value of 20% unfavorable change . . ............................................... $ (161)
Retained Interests, Guarantee Asset (Single-Family Only)
Weighted average IRR assumptions ............................................................ 8.5%
Impact on fair value of 100 bps unfavorable change ............................................... $ (382)
Impact on fair value of 200 bps unfavorable change ............................................... $ (714)
Weighted average prepayment rate assumptions ................................................... 20.1%
Impact on fair value of 10% unfavorable change . . ............................................... $ (517)
Impact on fair value of 20% unfavorable change . . ............................................... $ (995)
We receive proceeds in securitizations accounted for as sales for those securities sold to third parties. Subsequent to
these securitizations, we receive cash flows related to interest income and repayment of principal on the securities we retain
for investment. Regardless of whether our issued mortgage-related security is sold to third parties or held by us for
investment, we are obligated to make cash payments to acquire foreclosed properties and certain delinquent or impaired
mortgages under our financial guarantees. Table 11.4 summarizes cash flows on retained interests related to securitizations
accounted for as sales during 2009 and 2008. Cash flows associated with our retained interests in 2010 were not significant.
230 Freddie Mac