Freddie Mac 2005 Annual Report Download - page 127

Download and view the complete annual report

Please find page 127 of the 2005 Freddie Mac annual report below. You can navigate through the pages in the report by either clicking on the pages listed below, or by using the keyword search tool below to find specific information within the annual report.

Page out of 171

  • 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
  • 159
  • 160
  • 161
  • 162
  • 163
  • 164
  • 165
  • 166
  • 167
  • 168
  • 169
  • 170
  • 171

For periods prior to January 1, 2005, the Guarantee obligation fair value was calculated using internal models to
estimate future cash Öows using a Monte Carlo simulation. The components of estimated cash Öows associated with the
Guarantee obligation included estimates of expected future credit losses using statistically based models that were
benchmarked periodically to the non-conforming loan, or jumbo, securitization market. For all periods our estimates
included costs to administer the collection and distribution of payments on the mortgage loans underlying a PC and
considered net cash Öows due to security program cycles.
Recognized PC residuals
The fair value of recognized PC residuals is determined in a manner that is consistent with the approach described
above for the recognized Guarantee asset and Guarantee obligation.
Key assumptions used in the valuation of the Guarantee asset
Table 2.2 summarizes the key assumptions associated with the fair value measurements of the recognized Guarantee
asset. The assumptions included in this table for 2004 and 2003 relate to those used in our internal models to measure the
fair value of the Guarantee asset for single-family loans at the time of securitization and the subsequent fair value
measurements, which occurred throughout each year. For 2005, the fair values at the time of securitization and the
subsequent fair value measurements were estimated using third party information. The assumptions included in this table for
2005 are those implied by our fair value estimates, with the Internal Rates of Return, or IRRs, adjusted where necessary to
align our internal models with estimated fair values determined using third party information. Prepayment rates are
presented as implied by our internal models and have not been similarly adjusted.
Table 2.2 Ì Key Assumptions Utilized in Fair Value Measurements of the Guarantee Asset
2005 2004 2003
Valuation Assumptions for the Guarantee Asset Range(3) Mean(4) Range(3) Mean(4) Range(3) Mean(4)
Internal rates of return(1) ÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏ 1.8% - 13.8% 8.7% (1.4)% - 13.6% 6.7% 4.5% - 15.1% 9.4%
Prepayment rates(2)ÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏ 7.6% - 59.8% 17.2% 6.9% - 58.6% 19.1% 8.0% - 62.9% 23.3%
(1) The IRRs reported above represent a duration-weighted average of the discount rates used to value the recognized Guarantee asset. In 2004 and 2003,
such rates were derived by determining a single rate that equated (a) the simple average of future cash Öows of the Guarantee asset for each Loan
Group with that of (b) the calculated fair value of the Guarantee asset for each Loan Group. In 2005, the IRRs represent a duration-weighted
average of the discount rates as adjusted to align with our estimated fair values. Negative IRRs can occur when suÇciently large negative option-
adjusted spreads are applied to LIBOR. When we calibrate our modeled discounted cash Öows to the traded price of an IO security, a negative option-
adjusted spread can result when the traded price exceeds the implied market value of the modeled discounted cash Öows. A negative option-adjusted
spread is necessary to calibrate the implied market value of the modeled discounted cash Öows to the traded price.
(2) Average Prepayment rates are simulated on a monthly frequency, although rates reported above represent an unpaid principal balance weighted
average of annualized values of such Prepayment rates.
(3) The lowest value in each presented range represents the Ñrst percentile IRRs and prepayment rates throughout 2005, 2004 and 2003. Likewise, the
highest value in each range represents the 99th percentile IRRs and prepayment rates throughout 2005, 2004 and 2003.
(4) Reported values represent the weighted average value of all IRRs and prepayment rates throughout the 2005, 2004 and 2003 periods.
Weighted average lives of the Guarantee asset during 2005, 2004 and 2003 ranged between 1.6 and 8.9 years, 1.2 and
8.7 years, and 1.0 and 8.5 years, respectively, while the average derived weighted average lives of the Guarantee asset for the
same periods were 5.1, 5.3 and 4.8 years, respectively. Such derived weighted average lives are reÖective of prepayment
speed assumptions cited in Table 2.2 above.
At December 31, 2005, the fair value of the recognized Guarantee asset was based upon a valuation approach that
incorporates market-based information. In order to report the hypothetical sensitivity of the carrying value of the Guarantee
asset to changes in key assumptions, we used internal models to approximate their reported carrying values. We then
measured the hypothetical impact of changes in key assumptions using our models to estimate the potential view of fair
value the market might have in response to those changes. In our models, the assumed Internal Rates of Return were
adjusted to calibrate our model results with the reported carrying value. The sensitivity analysis in Table 2.3 illustrates
hypothetical adverse changes in the fair value of the Guarantee asset for changes in key assumptions at December 31, 2005.
Table 2.3 Ì Sensitivity Analysis of the Guarantee Asset
December 31, 2005
GA(1)
(dollars in millions)
Fair valueÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏ $4,938
Weighted average IRR assumptions: ÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏ 9.2%
Impact on fair value of 100 bps upward change ÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏ $ (184)
Impact on fair value of 200 bps upward change ÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏ $ (354)
Weighted average prepayment rate assumptions: ÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏ 15.5%
Impact on fair value of 10% upward change ÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏ $ (212)
Impact on fair value of 20% upward change ÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏÏ $ (404)
(1) At December 31, 2005, our Guarantee asset totaled $5,083 million on our consolidated balance sheet and of that amount, approximately $145 million
(or approximately 3 percent), relates to PCs backed by multifamily mortgage loans. The sensitivity analysis presented in Table 2.3 relates solely to the
Guarantee asset associated with PCs backed by single-family mortgage loans.
111 Freddie Mac