The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
Special Comment
July 2008
Table of Contents:
Highlights
4
Introduction
Accuracy Ratios
5
Investment-Grade Loss Rates
7
Average Rating Before Impairment
9
Rating Action Rates and Large Rating Action
Rates
11
Appendix: Description of Data Sample and
Glossary
16
Related Research
20
The Performance of
Structured Finance Ratings:
Full-Year 2007 Report
Highlights
This Special Comment updates Moody's structured finance rating performance
metrics as of December 2007. The highlights of this report are:
•
Overall, 2,090 structured finance securities became impaired in 2007:
20 in US ABS excluding HEL, 1,384 in US HEL, 112 in US RMBS, 13 in
US CMBS, 537 in global COOs, one in the international structured
finance sector excluding COOs and Other SF, and 21 in Other SF. Of
these, 1,780 were principal impairments (suffered principal losses or
were downgraded to Ca or C), while the remaining 310 were interest
impairments (experienced interest shortfalls only).
•
The one-year accuracy ratio for the most recent cohort was 77.2%,
which is 13.1 percentage pOints lower than its six months-prior level and
1
9.0 percentage points lower than the historical average. The decline in
performance can be attributed to the US housing crisis and its negative
impact on securities with exposure to US subprime mortgages.
•
The five-year accuracy ratio, which is a lagged indicator of performance,
was 80.1 %, 3.3 percentage points higher than its level of 76.8% six
months ago.
•
The one-year investment-grade loss rate jumped to 0.59% for the cohort
ending December 2007, a historical high and a more than five-fold
increase from the historical average of 0.11 %.
Analyst Contacts:
New York
1.212.553.1653
Julia Tung
Vice President - Senior Credit Officer
Nicolas Weill
Structured Finance Chief Credit Officer
1 These performance metrics should be interpreted with caution. Some statistics are based on small samples, as the number of impairments in any
given year and any given sector is often small. Individual performance metrics may also be driven by highly correlated collateral performance
underlying multiple securitizations, particularly within certain sectors. Moreover, variations in rating performance over time may reflect either changes
in the quality of the rating process or changes in the environment that make losses more or less difficult to predict or make collateral performance more
volatile, thus resulting in either higher rating accuracy and lower rating stability at one time than another, or vice versa.
Moody's Investors Service
•
The average rating during the three years prior to impairment of all securities that were impaired for the
cohort ending in December 2007 rose to Baa3, three notches above the average rating prior to impairment
for the cohort ending six months earlier, and a notch above the historical average.
•
The one-year rating action rate more than doubled in the most recent cohort to 9.6% compared to its six
months-prior level. Meanwhile, the frequency of large rating actions (rating changes of three notches or
more) more than tripled over the same time period. Both increases were caused by an upsurge in the
downgrade rate, particularly for US HEL.
•
Performance was mixed among the various sectors of structured finance (Figure 2). US HEL, US RMBS,
and global COOs all experienced lower than average one-year accuracy ratios and higher than average
one-year investment-grade loss rates compared to the historical experience. The reverse was true for US
ABS excluding HEL, US CMBS, and the international structured finance sector excluding COOs and Other
SF, all of which exhibited one-year accuracy ratios above 95% and one-year investment-grade loss rates
at or below 0.1 %.
Figure 1 - Summary of All Structured Finance Rating Performance as of
December 20072
Cohort Ending Date
December 2007
Number
of New
Impairments
over
Prior 12
Months
1-Year
Accuracy
Ratio
5-Year
Accuracy
Ratio
1-Year
InvestmentGrade
Loss Rate
36MonthAverage
Rating
Before
Impairment
1-Year
Rating
Action
Rate
1-Year
Large
Rating
Action
Rate
2090
77.2%
80.1%
0.59%
Baa3
9.6%
6.4%
June 2007
187
90.3%
76.8%
0.04%
Ba3
4.6%
2.1%
December 2006
108
96.4%
78.4%
0.01%
B2
4.8%
2.1%
June 2006
90
97.2%
79.7%
0.00%
B2
4.2%
1.7%
Average (1993-Most Recent)
220
86.2%
84.1%
0.11%
Ba1
4.9%
2.4%
2 A glossary appears at the end of this report. The number of impairments for historical cohorts is subject to revision during each update as payment:
shortfalls can be cured and past remittance or trustee reports may be revised. In addition, consistent with Moody's annual default and loss study,!
Moody's now derives loss rates using loss-given-default (LGD) from principal impaired securities alone. For more details, see the entry for LGD in the!
glossary. The historical average of the number of new impairments over the prior 12 months is calculated as the total number of newly impaired tranches:
divided by the number of years in the sample period, and has been rounded to the nearest integer unless rounding results in zero.
.
::::::::::::::::::::{JUly 2008 • Special Comment. Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Figure 2 - Summary of Structured Finance Rating Performance by Sector as of
December 2007
5-Year
Accuracy
Ratio
1-Year
InvestmentGrade
Loss Rate
36MonthAverage
Rating
Before
Impairment
1-Year
Rating
Action
Rate
1-Year
Large
Rating
Action
Rate
96.2%
82.2%
0.01%
B3
2.9%
0.9%
14.9%
Number
of New
Impairments
over
Prior 12
Months
1-Year
Accuracy
Ratio
20
Cohorts Ending December 2007
US ABS ex HEL
US HEL (includes subprime)
1384
74.7%
88.0%
1.81%
Ba1
19.1%
US RMBS (includes Alt·A)
112
93.3%
90.9%
0.06%
Ba1
5.4%
3.5%
US CMBS
13
95.7%
90.4%
0.00%
B3
11.0%
3.4%
Global COOs
537
61.7%
61.8%
0.66%
Baa2
9.9%
6.8%
3
99.9%
95.3%
0.00%
Caa3
3.4%
1.1%
21
57.5%
73.3%
2.83%
Baa1
7.0%
5.8%
US ABS ex HEL
33
86.2%
80.2%
0.22%
Ba1
5.7%
3.3%
US HEL (includes subprime)
108
88.5%
91.4%
0.19%
Ba1
3.8%
2.6%
US RMBS (includes Alt·A)
11
94.8%
96.1%
0.01%
Ba2
2.8%
1.1%
US CMBS
8
92.5%
87.9%
0.01%
B3
12.3%
4.2%
Global COOs
58
Int'l SF ex COO ft Other SF
Other SF
Historical Averages Since 1993
76.8%
64.0%
0.33%
Baa3
8.6%
4.6%
Int'l SF ex COO ft Other SF
70.4%
83.2%
0.01%
Ba1
4.1%
1.3%
Other SF
59.3%
81.5%
0.09%
Baa1
1.7%
0.3%
..........................................................................................:.......................................................................................................:....................................................
:::::::::1i:;:::::::{JUly 2008 • Special Comment. Moody's Credit Policy - The Performance of Structured Finance Ratings: Full·Year 2007 Report
Introduction
In a Special Comment published in April 2003, Moody's developed a set of metrics to measure the
performance of corporate ratings with respect to the dual objectives of rating accuracy and rating stability.3
4
Moody's corporate rating performance report is now updated on a quarterly basis Moody's first introduced
and examined its structured finance rating performance metrics in a September 2004 Special Comment
"Default & Loss Rates of Structured Finance Securities: 1993-2003," and published these performance metrics
in a stand-alone document for the first time in September 2005. The structured finance rating performance
report is now updated on a semi-annual basis.
For both the corporate and structured finance rating performance reports, the basic unit of observation is a
monthly cohort of ratings, i.e. all outstanding ratings at the beginning of a month are recorded and their
performance tracked over different time horizons. In computing rating performance metrics for structured
finance, Moody's incorporates both the default and loss severity experience of all structured finance tranches
because Moody's structured finance ratings rank order expected loss rates. In other words, Moody's
structured rating performance metrics weigh those tranches that have become materially impaired but with
lower loss severity less than those with higher loss severity.
The most important measure of rating accuracy is the accuracy ratio, which for structured finance measures
the relationship between tranche ratings and their realized loss rates.s This metric measures the quality of
Moody's ratings as indicators of relative expected credit loss risk.
As discussed in the April 2003 Special Comment, although relative rating accuracy is our primary objective,
Moody's recognizes that many investors are also concerned with the cardinal accuracy of the rating system.
In particular, they expect that investment-grade credits should rarely suffer credit losses and impaired
securities should normally carry low ratings well in advance of impairment. For this purpose, we regularly
track investment-grade loss rates and the average rating of securities during the 36 months prior to
impairment. Both of these measures should be low if the rating system is accurate in a cardinal sense.
We employ two measures of rating stability (or rating volatility) - the rating action rate (the frequency of rating
changes) and the large rating action rate (the frequency of rating changes of three notches or more) over a
twelve-month period. To examine how watchlist actions are used, we also report the percentages of
downgrades and upgrades preceded by watchlist (review) actions in the same direction.
Note that the criteria used to create the data set for this report have changed from prior performance studies.
The most notable changes are that pari-passu tranches are no longer collapsed and wrapped tranches are
included. For a more detailed description of the data sample, please see the Appendix.
See "Measuring the Performance of Corporate Bond Ratings," Moody's Special Comment, April 2003.
For the latest performance report, see "The Performance of Moody's Corporate Bond Ratings: March 2008 Quarterly Update," Moody's Special
Comment, May 2008.
~
5 The required adjustments to convert the standard default-based AR measure to a loss-based measure are discussed in "Default & Loss Rates of
Structured Finance Securities: 1993-2003," Moody's Special Comment, September 2004. The concept of accuracy ratio is also described in the glossary:
at the end of this report.
3
4
:::::::::t:::::::fJUly 2008 • Special Comment • Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Accuracy Ratios
Figure 3 depicts the one- and five-year accuracy ratios over time with the following notable observations:
•
For global structured finance, the one-year accuracy ratio decreased to 77.2% from its level six months
ago of 90.3% due to an increase in the number of material impairments in the second half of 2007 within
the US HEL sector, composed mostly of securities backed by subprime mortgages, and within global
COOs, composed mostly of securities backed by a portfolio of other structured finance securities.
•
Global COOs experienced the steepest decline in the one-year accuracy ratio, dropping to 61.7% for the
cohort ending December 2007, a 32 percentage point decrease from its level six months prior, and a 35
percentage point decrease from its level 12 months ago. The one-year accuracy ratio for US HEL also fell
dramatically to 74.7%, a 15 percentage point decline from the cohort ending June 2007 and a 18
percentage point decline from the cohort ending December 2006.
•
In contrast, the one-year accuracy ratios for US ABS excluding HEL, US RMBS, US CMBS, and
international structured finance excluding COOs and Other SF were essentially unchanged from their six
months-prior levels and were all above 93%.
•
Unlike their one-year counterparts, there were no dramatic movements in the five-year accuracy ratios
between the July 2002 cohort and the January 2003 cohort for global structured finance or any of its subsectors. This statistic demonstrates performance on a lagged basis and the effects of the recent dramatic
increase in material impairments have not been incorporated yet.
::::::::::::;::::::::f" July 2008 • Special Comment • Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Figure 3 - One-Year (yellow line) and Five-Year (blue line) Accuracy Ratios
USABS ex HEL
Global Structured Finance
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
~':':,;;.'".,.,.,:.::::':...., ..•;:.,,;... :,:.......:........~..~.;~.-,' .~.~
..~.,~'~"
.=-".,;,..- . .
~--~
+---------------------------------------+---------------------------------------+---------------------------------------+---------------------------------------+---------------------------------------+----------------------------------------
ffi
ill
i35
8l
'"
-,
'"
-,
'"
-,
c
c
-,
c
ac
c
'"
8
c
'"
'"
-,
-,
iB
t;
'"
-,
c
-,
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
~
c
'"
'"
-,
~
'"
-,
Cohort Starting Date
us
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Cohort Starting Date
US RMBS
HEL
,------
~.....:...:.»,""'~".:~~:.,~.;.:...:,,::c:~~.:-
+---------------------------------------~
+---------------------------------------~
+---------------------------------------~
+---------------------------------------~
ill
ffi
-,~
-,~
a
8l
i35
-,~
-,~
8
-,~
-,~
iB
-,~
100%~ ..,.,.,.,................ ,., .......,..:.......:: .. :, ...........".
90% - " - " - :
... '-;;.
80%
70%
60%
50%
40%
30%
20%
10%
0%
t;
-,~
EE
C7
••
c-;.
k:::;:::::?.·!···· .....
~~~ +-'~~~·<~~L'__________________________________~
20%
10%
0%
ffi
-,~
ill
-,~
i35
-,~
a
8l
-,~
ffi
8
-,~
-,~
8
...,§
-,~
iB
§
-,
t;
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
+-------~------.----
+-_____ .)".-..-.:.. - -•.:c .......-_.;_-;-._,-
"-~;-~
.••-...-..- .. - . ~~·--=------·:7:·---::~
ffi
ill
'"
-,
c
§
-,
-,
c
'"
Cohort Starting Date
Cohort Starting Date
Int'l SF ex COO & other SF
1~~: ,------------c_:;:co;~:;;;.;::........... j:
80%
60%
50%
40%
30%
20%
10%
0%
Other SF
,.: ".---:----.
--~.---~
....... ,:
,!-'
~~
m%
·.~
Global COOs
=.. c-.c-...=. :.>. ~.:. ;~. ~. '-;.~.,~.--.~. ~. .-...
00'
__
Cohort Starting Date
,---~~---------~---~----------~~---
1----J--------":'''.. =.
. . . ....
c·~·
~
USCMBS
ri.
..
+-------------------------------------+-------------------------------------+-------------------------------------+-------------------------------------+-------------------------------------+-------------------------------------_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
Cohort Starting Date
100 0k
90%
".
+--------------------------------------
~,----------------
+---------------------------------------+---------------------------------------+---------------------------------------+---------------------------------------+---------------------------------------+-___- _______ ____ __
~
~
~
~
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
ffi
ill
i35
8l
'"
-,
'"
-,
'"
-,
c
-,
-,
Cohort Starting Date
c
c
c
'"
ac
'"
-,
8
iB
t;
'"
-,
'"
-,
c
-,
c
c
'"
Cohort Starting Date
Note: At the beginning of each month, securities are grouped together by their alpha-numeric ratings to form rating cohorts. The
latest one- and five-year cohorts are formed on January 1, 2007 and January 1,2003. Breaks in the accuracy ratio series occur
when the number of impairments for the cohort is less than or equal to one. Crosses in the one-year series and plus signs in
the five-year series indicate that the accuracy ratio was computed from only one impaired security.
:::::::t::::::::::f July 2008
•
Special Comment •
Moody's Credit Policy
~ The Performance of Structured Finance Ratings: Full-Year 2007 Report
Investment-Grade Loss Rates
Figure 4 shows one-year and five-year investment-grade loss rates by sector. Note that:
!!!!!!!!!!:~:!!!!!!fJUIY 2008 •
•
For global structured finance, the one-year investment-grade loss rate reached an all-time high of 0.59% for
the cohort ending December 2007, up precipitously from 0.04% six months prior and 5 times higher than the
historical average of O. 11 %.
•
The one-year investment-grade loss rate also rose to record-breaking highs for US HEL and RMBS,
increasing to an unprecedented 1.8% for HEL and 0.1 % for RMBS. The loss rate for global COOs also
increased dramatically, but was still lower than the peak of 2.1 % for the cohort formed in May 2001.
•
For the cohort formed in January 2007, the one-year investment-grade loss rate was zero for US CMBS
and the international structured finance sector excluding COOs and Other SF as it has been for the last 2
years. For US ABS excluding HEL, the rate was a very low 0.01%.
•
The five-year investment-grade loss rate for US HEL, US RMBS, and the Other SF sector all showed a
pattern of increase over the last six months. For all other sectors, including overall structured finance, the
five-year investment-grade loss rate exhibited a declining trend.
Special Comment •
Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year
2007 Report
Figure 4 - One-Year (yellow line) and Five-Year (blue line) Investment-Grade Loss Rates
Global Structured Finance
us ABS ex I-EL
3.0% - , - - - - - - - - - - - - - - - - - - - - -
3.0% - , - - - - - - - - - - - - - - - - - - - - -
+-_________,::,~,:,.::C'..:."-::.--"-::.,-----
2.5%
1.0%
+-------------------+-------------------+-------------------+------------;..0...···~··O;_-----
1.0%
+-------,:'".::Liii:-·:------.c----
0.5%
+-_____----,.....::.~. L.·~---------
0.5%
+-------,-------c:-.. - . - - - ......
2.5%
200k
1.5%
..
, ;.
~::: :==================:i~r_~'_'_________
Cohort Starting Cl3te
Cohort Starting Cl3te
us I-EL
us RMBS
3.0% - , - - - - - - - - - - - - - - - - - - - - - - -
0.10% - , - - - - - - - - - - - - - - - - - - - - -
2.5%
+.---------------------
~f~\--,-··~::-------------
2.0%
0.08%
+--------------------
0.06%
+--------------------
0.04%
+-------------r'.'. . -----.
0.02%
r--t·······..·:,~:~c--------i---"""C""-~
~~:~~c~~:~·---------------:
1.5%
..::/
'
1.0%
0.5%
+------i-~------------.
0.0%
-I-~~,..;.
···\.=. .:......-......--:....--:-......cc::......=..... -----:.. :: ..... "':.:'" ..ct· ...
__~,..;.~,..;.~_=_===-,-;.~
ii
.....
:.......:.:.::: ... .
0.00% +-....,.:.~~'c-c===~="~'¥
.....,.i~=_~='"-
~ ~
~ ~
Cohort Starting Cl3te
Cohort Starting Cl3te
us CMBS
Global COOs
1.2% - , - - - - - - - - - - - - - - - - - - - - - -
6.0% - , - - - - - - - - - - - - - - - - - - - - -
1.0%
5.0%
+.~--------------------
0.8%
4.0%
....,
+--------:;--7 ~.."':.:',-- - - - - - +-_____~~:'-\-,..-\-f---'~,~------
0.6%
iJ________________
3.0%
+-_____~:~0-~-'-----:~~-------
0.4%
!\:;-:--------,,~.~------
2.0%
+------.:c------ .._·:~t,,~-----
:--------------
1.0%
+-------;",-""'----"C
>:
i~
0.2%
r'
;. .: . . . . ;
......
Cohort Starting Cl3te
Cohort Starting Cl3te
Int'l SF ex COO & other SF
other SF
0.20% - , - - - - - - - - - - - - - - - - - - - - -
3.0% - , - - - - - - - - - - - - - - - - - - - - -
0.16%
0.12%
0.08%
+-----------::.::'c:---------
2.5%
+--------------------.:
2.0%
+ - - - - - - - - - - - - - - - - - - - - :•
1.5%
1.0%
+--------------------:
+---------------------'
0.5%
+ - - - - - - - - - - - - - - y - - - - - -•
<>~:
+-----------.::-.:;.-------
+-----------c:p....,.
0.04%
+---------~-.-~~------
0.00%
+---______
.' t. .
~~-~~~~
___
~
...,
Cohort Starting Cl3te
Cohort Starting Cl3te
Note: At the beginning of each month, all securities carrying an investment-grade rating are grouped together to form a
rating cohort. The latest one- and five-year cohorts are formed on January 1, 2007 and January 1, 2002.
:::::::::;;::;::::::{ July 2008 •
Special Comment •
Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Average Rating Before Impairment
Figure 5 presents the 36-month-average rating before impairment over time as well as the number of newly
impaired securities used to calculate the average ratings. The following observations are noteworthy:
iiiiiiiii:::Jiiii{July 2008 •
•
The 36-month average rating before impairment for global structured finance rose to Baa3 for the most
recent cohort, three notches up from Ba3 six months prior, and one notch above its historical average of
Ba1. The increase can be attributed to increases for US HEL, US RMBS, and global COOs.
•
The average rating prior to impairment for global COOs increased six notches to Baa2 for the cohort
ending December 2007 from B2 for the cohort ending June 2007. Those of US HEL and US RMBS rose
one notch and two notches, respectively, over the same time period.
•
The average ratings before impairment for US ABS excluding HEL, US CMBS, and international structured
finance excluding COOs and Other SF were unchanged from their levels six months prior at B3, B3, and
Caa3, respectively. In addition, all were at or below their historical averages.
Special Comment •
Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Figure 5 - 36-Month-Average Ratings before Impairment (yellow line) and
Number of Newly Impaired Securities (blue line)
Global Structured Finance
Aa3
~----------------~
2500
A3
~---------------;~
2000
::
us ABS ex HEI...
Aa3 ,---------------------------------,200
A3
Baa3
Ba3
B3
Caa3
bl
C:
...,'"
~
...,'"
Cohort Starting Date
,....
OJ
C:
...,'"
Cohort Starting Date
USHEI...
Aa3,--------------------------------C! 1500
Aa3
~----------------__,
150
+------------------,'f
A3
+-------------------+
120
A3
Baa3
Ba3
+----------,
1200
USRMBS
c - - - - - - - - - : t 900
~.,.,.,.~~..... ~.''--'- - - c-c-:'---".. '..-'-....-...'.-"--'/ - - - - -...-.,"4': 600
B3 + - - - - - - - - - - - - - - - - - i + 300
Caa3 -!-,~~~_ _~_ _~'F""~_ _ _i"'c''+~j 0
B
a a 3 L : --_ _ _ --_______ - - - - _.::[' 60
90
Ba3
B3
Caa3
...,~
=""
~
J
~
J
!
" J ' " " ,,"""" " " " , ' '"
~
J
USCMBS
,/,
';
.
.(, ' , , , ' , , , ' "
0
~
J
~ 500
A3
i-
Baa3
375
Ba3
B3
.........
C:
...,'"
~
...,'"
,....
OJ
C:
...,'"
~
...,'"
I'W
"""""""
/.-,0"
Caa3
bl
aC:
...,'"
,
8
C:
...,'"
'!r
125
0
.. ;'.""":" ..'i'
,....
~
0
C:
...,'"
...,'"
Cohort Starting Date
Inti SF ex COO and Othe r SF
Aa3 ,----------------------------------, 20
Other SF
50
Aa3
16
A3
40
+--------------'··--412
Baa3
30
+-------------------~:------------_+
Ba3 +---------------------------------_+ 8
.:
B3
~
J
625
Cohort Starting Date
A3
~
J
Aa3
...,~
Baa3
~
J
Global COOs
~: = =" = = = = = = = = = = =~ -)=-i_ =:f-:-;=j-} \-'~ 'il\;~r' ~
Caa3
0
Cohort Starting Date
Cohort Starting Date
......,.
~
J
30
, , " " ",,,,,' ,,,,,,,,,
..'
Ba3
.f 20
i: 10
+----------+--+------+
4
B3
i""j""\,i"":Si
0
Caa3
Caa3
....;
...,~
bl
...,~
Cohort Starting Date
...,~
...,~
bl
C:
...,'"
~
...,
'"
,....
OJ
C:
...,'"
~
...,
'"
aC:
...,'"
8
C:
...,'"
~
...,
'"
,.... 0
0
C:
...,'"
Cohort Starting Date
Note: At the beginning of each month, all securities that become impaired within the next 12-month period are grouped to
form a cohort, The latest 12-month cohort is formed on January 1, 2007, Breaks in the average rating before impairment
series occur when the number of impairments for the cohort is less than or equal to one, Crosses indicate that the average
rating was computed from only one impaired security.
iiiiiii:;jti:;iii{ July 2008
•
Special Comment •
Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Rating Action Rates and Large Rating Action Rates
Figure 6 reports 12-month rating action rates and large rating action rates. Figures 7 and 8 further
disaggregate rating actions into downgrades and upgrades, and Figure 9 demonstrates how frequently
6
downgrades and upgrades have been preceded by watchlist actions in the same direction. Key observations
include:
•
The 12-month rating action rate in the global structured finance category was 9.6% for the January 2007
cohort, up roughly twofold from the rate of 4.6% six months prior and the historical level of 4.9%. The
large rating action rate also increased significantly to 6.4%, from 2.1 % six months ago and 2.4%
historically.
•
The global structured finance downgrade rate increased dramatically to a historical high of 7.4% from
1.4% six months earlier. The large downgrade rate also rose from 0.8% to 5.7% over the same
period. The major contributor of structured finance downgrades was US HEL. In fact, US HEL
experienced the largest increase in the downgrade rate and in the frequency of large
downgrades. Downgrade rates for US RMBS, global COOs, and Other SF also experienced large
increases in the latter half of 2007. In contrast, US ABS excluding HEL, US CMBS, and the international
structured finance excluding COOs and Other SF exhibited frequencies of downgrades that were low both
in absolute and historical terms.
•
The overall upgrade rate for the most recent cohort was 2.2%, a decrease from both the upgrade rate of
3.2% six months ago and the rate of 3.6% twelve months ago. The frequency of large upgrades also
experienced a corresponding decline of 0.7% from 1.3% six months prior and 1.5% twelve months
prior. All the sectors of structured finance, with the exception of the Other SF category, displayed a
declining trend in the frequency of upgrades.
•
The proportion of downgrades and upgrades that were placed on review prior to the rating action fell to
22% and 23%, respectively, for the cohort ending December 2007. The decline in the overall percentage
of reviewed downgrades was led by corresponding decreases among US HEL and US RMBS, where the
percentage fell to 19% and 9%, respectively. The proportion of upgrades reviewed before the rating action
decreased for most sectors, falling below 4% for US ABS excluding HEL and below 1% for US CMBS. US
HEL and global COOs were the only two sectors for which the percentage of reviewed upgrades remained
relatively high at approximately 75%.
Moody's also reviewed its uses of watchlist actions on structured finance securities in a Special Comment, "Structured Finance Watchlist Resolutions:
1992-2003," June 2004.
6
iiiiiii:;:iii:;:iiii{July 2008 • Special Comment • Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Figure 6 - 12-month Rating Action Rates (yellow line) and 12-month Large
(three notches or more) Rating Action Rates (blue line)
Global Structured Finance
USABSex HB...
180/0
160/0
140/0
12%
10%
6%
4%
+--------------------------------------+-------------------------~
---------
~!~ +-------------------------;~~tr::r~ :
8%
2%
~--------------------------------------
..
I
1"""·"""",:",,,/ ..
....
ffi
t:
~
t:
-,
-,
-,
;:.~./
0% I'
'"
~~:
-:". . . _,,__ ", . . :. . _~: . . ·,·,·~~C,~,~·~·~,~·~""
00/0 +l~;~,,__~~__~~.:~~"-'_~~'~v~·~_"~v~,"-"----__- - - -__-·~'
~
(J)
'"
....l
.....:-..........:...
'"
is
ffi
'"
-,
t:
-,
'"
-,
8
§
-,
§
Cohort Starting Dote
Cohort Starting Dote
USHB...
250/0
USRMBS
~--------------------------------------
20%
15%
100/0
+---------------------------------------
5% t --------. c .
------------------------:.~jJ
..~
.:)"<.\. T······;{······
0%
~
-,
Cohort Starting Dote
Cohort Starting Dote
USCMBS
Global COOs
250/0 ,---------------------------------------
20%
16%
20%
15% +-----------------------:-A
. :-: :--:.-".----.._.- - - - - - - - - - - - - - - - - -
12%
:"/'
M~h",. u-:~,._._., _.
8%
10%
- - c - - - - - - y....
. """.,/
4%
0%
I.}
\:y.. : ':.-,.'
- - - - - - -..
.-------.---"- ·~·~·-~-t·~"---':~~.-:~:
5%
00/0
~
":
"f
.""\:
••••••.••••••••.••••••• :.>.•.•••.•
'~.
-.---.-'";'
.. ~~/ t ......, ........
.; .......; ";/
+-~--~--~~,-~----
________________"'
~
-,
-,
Cohort Starting Dote
Cohort Starting Dote
Int'l SF ex COO & other SF
other SF
20%
12% I
100/0 t·-------------------------------------
16%
12%
8%
6%
t"';: :':.....;:.
:
8%
............
4%
.....
.------
00/0 +I-·~-__~,~.,~,~+_~,_·~~~~~~~,-~~,,~;~;-·;·-~--~-~·-~·_·~_"_"_'__"_'._.~-"_"_~~"
~
4%
:
0%
~::
~
-,
J
Cohort Starting Dote
:
:
....... :.
::
:.:
:..
.-:
:
::
. : . : : . : .....
2%:C~"~'_:~-.-.- : - : - .. :.--.--.-::~:
: . : ... :::
~
J
1
J
<: :.. : tl. :
~
J
~
J
. .1
8
t:
-,
'"
~
'"
-,
....0
t:
-,
'"
Cohort Starting Dote
Note: Rating actions include upgrades and downgrades, which are measured on the alpha-numeric (or modified) rating
scale. Rating cohorts are formed each month covering a 12-month period. The latest 12-month cohort is formed on January
1, 2007.
::::::::;:lI:{ July 2008
•
Special Comment •
Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Figure 7 - 12-month Downgrade Rates (yellow line) and 12-month Large
(three notches or more) Downgrade Rates (blue line)
USABSex HEI...
Global Structured Finance
16%
14%
12%
10%
8%
6%
4%
2%
+-----------r:\
-. . ..
+-----~
~'-.. --;.~:'--~.j::.........
......... .;;.'
0% I ........
5l
'.'
iJ;
i5l
c
...,'"
c
...,'"
Sl
c
...,'"
Cohort Starting Date
...,~
8
c
...,'"
is
c
...,'"
is
c
...,'"
t;
c
...,'"
Cohort Starting Date
us HEL
US RMBS
20%
16%
12%
8%
•.....
".
4%
/;"\~:
t·'···il·;;;\,,;;;:.
;;;}
0%
...,~
Cohort Starting Date
Cohort Starting Date
us CMBS
Global COOs
14%
~--------------------------------------
12%
+------.~--------------------------------
10%
t:'2,.~ __---------------------------
8%
:~2 ;.: :!l:'~: : :.:.[..• ~-_-_-_-_-_-_-_-_-_-_-_-_-~.:.
•.
ro
0%
.:-
. '-'---.'-"-.---
....:•.'M"!;.:~ .. - . .-. -.. -.
..........
..•.;...:/..•.•. ....... ... __~__~~~~~
+-~~~--~:+~~~~~~~~
.'
~
Sl
...,~
...,
is
...,~
8
...,~
is
...,~
25%
~-----------------------------------
20%
+---------------------~.~ ~----------
15%
+----------------------..~/>;:_..----------
..
..
0%
t;
:)'::~.....
........
________
+-~~~~~~
8%
..
4%
ro
~
~~~
Other SF
;~:
~~
I
..
8%1":-::_'-
6%
20 00:
~......... .......
__ ______
Cohort Starting Date
Int'l SF ex COO & Other SF
10%
~
~
...,~
~-------------------------------------
\
1:: :=================~-..-"'.-.,~l-: .\.-.. . ------~'", ,',.J.
Cohort Starting Date
12%
/
..
I
.. - . - - - - - - - - - - - - ..
.
..~
-~.----~-... --:~''''''-'.
. ," . ;.....
"": ... -:}~; •••::............... .
6%
--4 % - - : - - :----------------------------2%
.....
:--__---------------------------:
0%"
5l
...,~
Cohort Starting Date
i
.
i5l
Sl
...,~
...,~
is
...,~
8
...,~
Cohort Starting Date
Note: Downgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month
covering a 12-month period, The latest 12-month cohort is formed on January 1, 2007,
iiiiiii:;:ii!:i:;iii{ July 2008
•
Special Comment •
Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Figure 8 - 12-month Upgrade Rates (yellow line) and 12-month Large (three
notches or more) Upgrade Rates (blue line)
Global Structured Finance
USABSex HEI...
4%
::~\\
3%
...•.....
.
:: (=::~,2~:J~.···-/'::-····-·····-:··-····-··:-\.
~
~
"""?
~
"""?
"""?
~
~
"""?
"""?
;
2%
-
1%
0%
~
ffi
i5l
iJ;
ffi
Ei
'"
'"
'"
8
c
--,
'"
c
--,
--,
--,
--,
c
c
--,
"""?
Cohort Starting Date
c
c
'"
'"
Cohort Starting Date
us RMBS
USHEI...
7% ,---------------------------------------
6% t--------------------------------------5% t--------:-- .------- ...-------------"'-----4%
=_. __ : _:- ;. .;.-
t-----~---;------.;~~;-------~
~:
.-.------
~;- -:------L~~.:·.i:.:~;.............
..
1%
.::........<:.
.
.....
~ : :.-:.~~:~~.:-: . -...-::-------..~:~.
0% ~~._
... -.....;:".=~.,...;'-·~4~~-'·::;c.:'·.. ~···-"<·"')~~__~--,_
ffi
~
--,
Cohort Starting Date
us CMBS
Global COOs
t---------------------------------_c--'·~
t-------------------------------c;~~-
t - - - - - - : - ------------------:-----t-----------~c___;:" ,,;--------------;'------t----------.. ~~..
. ;-'. ~.-----...-...-.- - -..::'::;.~
.;L....:'i:.
-:--C-."
--;-o;r' ",(":";'\'
~•• - - - -••• £....-••. -.- - - - -
•••
+,__~~.~••~••~~~}_~~~__~__"_v_~-~~·___~~~
ffi
Ei
~
--,
~
--,
8
14%
12%
--.
10%
8%
6%
4%
2%
..
;
,
.........
N·;;;·.·.
',O',------T
"
i5l
iJ;
ffi
Ei
'"
--,
'"
--,
'"
--,
'"
--,
--,
c
c
c
Int'l SF ex COO & other SF
Ei
8
is
t;
'"
--,
'"
--,
'"
--,
Cohort Starting Date
8
c
is
t;
'"
--,
c
'"
--,
c
'"
--,
5%
4%
3%
:':
2%
•.. ; ?:
. :
1%
.:~.
c
'"
/
Other SF
7°k t-~-----------------------------------6%
5%
- - - - - - - - - - - :...- - . 4%
3%
2%..
;:.---.-'----.;-""\.. - 1%
. - . - - . - . - - -...=;:-_.~.~--.-....'.;~..,..
.;.-....•..;',' ",
0% -f....,---4-~~.Lt_··:;.···c:;·::cc.._·_;."'..;~:;~.";'~.:.'--;:::.~____'~__~'__,...__~____~
--,
c
.......'.
Cohort Starting Date
8% ,---------------------------------------
~
:;;.: ..>;:...... ;~; ;
ffi
c
~
--,
.... ..
'
0% ,"
Cohort Starting Date
--,
~
--,
Cohort Starting Date
,--------------------------------------
-----c- -
8
Ei
~
--,
18%
16%
14%
12%
10%
8%
6%
4%
2%
0%
:."" '::
c
c
c
'"
0%
ffi
i5l
iJ;
'"
--,
'"
--,
c
--,
c
ffi
'"
--,
c
c
'"
•• '
it
Ei
c
'"
--,
';.
;<
8
c
'"
--,
is
t;
'"
--,
c
--,
Cohort Starting Date
Note: Upgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month
covering a 12-month period. The latest 12-month cohort is formed on January 1, 2007.
:::::::::tJ::{ July 2008
•
Special Comment •
Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
c
'"
Figure 9 - Percentages of Downgrades (yellow line) and Upgrades (blue line)
Preceded by Watchlist Actions in the Same Direction
us ABS ex I-EL
Global Structured Finance
1000k
1000k
~--------~--------------------------
80%
+-----------:r:'.\:
80%
60%
'·,-,·cc·-·
60%
-!__ :' .
40%
20%
..
Oak
---------,f".;. . :\-.. . ..
~:.:
:;:
'
.....
;:
r-~lfl"--:~--'---'-'-----I ':.:
:~--::-J~l':
. .•~"---.----:.r-.:•. \~. . .: -.:.!...~. ~.- - - - - '
r--
~
--,
'"
us I-EL
.:.: ....:
.:.....:..
'"
100% ,_--------__
80%
/i
+----------
60%
.~._________
~.
~-P ~~~~ ~~~~~f~:~~~~~,
::,:-i__________
__
'"
--,
r--
~
--,
~
'"
'"
--,
us RMBS
- - - - - - - " - - - - _... "
0%
'"
--,
~
--,
~
Cohort Starting [ate
-~------~····7···············--
:::-.-..• _:(L.,
. •:.
~
--,
~
Cohort Starting [ate
60%
i , .-------:;
:)
.: ";"
.:.......
::;:
<;:
"1
':J
0% ~1·4-~/~·~~~~~~~~_~~~~
____~
~
80%
\
("
:, ;,
20%
;
+-~~~~~~~~~~-,~~~~~~~~
100%
~
: l:
40%
~,.:. : j·,~I~·,,~------,··<···.
....
:........ .. ....
,,:/
~-----y~_~:~--------------~--~~---
__
..: : ' : - - - - :...
:'
---i r
40%
;e-c-'-------l
20%:::
0%
~
'r
:'
"~.
~.~..~.,~(,~_~~~~:~-+;~~i~···~:~~_~i,.-~~~,__~___·~t
Sl
81
--,
--,
iii
--,
".. :~:....·::·::····f·--
. "r"
);.} ..
iii
Cohort Starting [ate
Cohort Starting [ate
Global COOs
USCMBS
100%
100%
[q.:FPU.::: . . \,.
l· l : :L;:,· ~; ...:\
Ie-iii .:i~.:•i- __.__l. ____~
80%
60%
40%
..
i1:-'-----:~:
20%
80%
60%
40%
: '(
:;-----------..
.;.:':.:"
'-'.
.... j ____.•.•.•.
+-~..~____~lL:~~~__~~~
~~~~~
20%
'
0%
~
~
'"
--,
--,
~
~
'"
'"
--,
--,
r--
~
0%
~
'"
Sl
i£
81
'"
'"
--,
.!:
--,
'"
--,
--,
Cohort Starting [ate
.!:
--,
iii
Cohort Starting [ate
Int'l SF ex COO & Other SF
100%
other SF
100% ,_-------
.---.,.l~,k----·~:-· ..
: i---rT ---:-<----~.
.... ", .
---f:-r:-:
..
.
.. ::...:.-- .~.-... ------'.-.--:;.-.
80%
60%
.
40%
_--.):;---------
.
··'l.
20%
0%
+-~~
__
II:....
~~~
__
i;\.:.\,.,....
~~~~~~~~
~
____
-r,
~~ +---=-·---~--·=~v~~
0%
~
--,
Cohort Starting [ate
Cohort Starting [ate
Note: At the beginning of each month, all securities that experienced downgrades (upgrades) within the next 12-month
period are grouped together to form a rating cohort. The latest 12-month cohort is formed on January 1, 2007. Gaps in the
data indicate that there were no downgrades (upgrades) during that time period.
::::::!::j::!:~:::::f July 2008
•
Special Comment •
Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Appendix: Description of Data Sample and Glossary
The data sample used in this report includes all public, 144A, and private tranches with a published Moody's
long-term global debt rating among global asset-backed securities (ABS), commercial and residential
mortgage-backed securities (CMBS and RMBS), collateralized debt obligations (COOs), and other structured
finance, including asset backed commercial paper (ABCP), structured investment vehicles (SIVs), structured
covered bonds, catastrophe bonds, and derivative product companies. Provisional ratings, credit estimates or
evaluations, short-term ratings, and national scale ratings are not included. The following types of securities
are excluded from the definition of global structured finance and therefore are not included in the data sample:
repackaged securities, structured notes, and other credit derivatives which are basically pass-throughs of the
rating of another entity.
This data set is an expansion of the data set that was used in prior structured finance performance studies?
Unlike the data set from previous years, this data sample:
•
Includes tranches wrapped by financial guarantors, government agencies, and government sponsored
enterprises (GSEs);
•
Includes interest-only (10) and residual tranches;
•
Includes some transactions outside of the four major sectors (ABS, COO, CMBS, RMBS) of structured finance,
such as ABCP, SIVs, structured covered bonds, catastrophe bonds and derivative product companies;
•
Does not collapse tranches with the same rating from the same deal, i.e. all pari-passu tranches are
counted in the data sample. The exceptions to this are notes with the same rating issued out of the same
program for ABCP, SIVs and structured covered bonds, in which case only the rating of the program and
not each individual security is counted.
The data used to create this report are commercially available via Moody's Structured Finance Default Risk service.
For more information, please email DefaultResearch@moodys.com.
Glossary
Material Impairment
Structured finance securities are defined as being in material impairment if they have:
•
Sustained a payment shortfall that remained uncured, or
•
Been downgraded to Ca or C.
Prepayment-related and AFC-related interest shortfalls are not considered to be material impairments, but
PIKing tranches are. Explicit principal write-downs are included whereas implicit principal write-downs or
under-collateralizations are not.
The impairment status of a security may change as it goes from cured (i.e. all outstanding shortfalls and losses
were repaid in full) to uncured (i.e. positive interest shortfalls or principal losses outstanding), or vice versa. If
any securities rated Ca or C but not in payment shortfall are upgraded, they are considered to be no longer in
material impairment. Securities rated Ca or C that were not upgraded are in material impairment even if their
payment shortfalls have been cured. Finally, securities with very minor shortfalls or losses are excluded.
The expanded data sample was first introduced in our 2007 structured finance rating transitions studies that were published this year. The data sample!
in this study was extracted following similar guidelines.
.
7
::::::::;jti:;:::{JUly 2008 • Special Comment • Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Payment Shortfall
Structured finance securities are defined as being in payment shortfall (previously called "payment default") if
they have suffered either one of the following:
•
Interest shortfall
•
Principal write-down.
Principal Impairment
This refers to materially impaired securities that have suffered principal write-downs or principal losses, or
have been downgraded to Ca or C even though a principal write-down or loss has not yet been observed. In
particular, if a security had experienced principal write-down/loss or was downgraded to Ca or C, it is called a
principal impairment regardless of whether it had experienced interest shortfalls.
Interest Impairment
This refers to materially impaired securities that have experienced only interest shortfalls, no principal losses,
and were not downgraded to Ca or C.
Investment-Grade (IG) and Speculative-Grade (SG) Ratings
Investment-grade ratings refer to Aaa, Aa1, Aa2, Aa3, A1, A2, A3, Baa1, Baa2, and Baa3. Below investmentgrade or speculative-grade ratings refer to Ba1, Ba2, Ba3, B1, B2, B3, Caa1, Caa2, Caa3, Ca, and C.
Loss Severity or LGD
The LGD rate of an impaired structured finance security is measured by the sum of the present values of net
losses, including both interest shortfalls and principal losses, discounted by the security's coupon rate and
expressed as a percentage of a given principal balance such as the principal balance at origination, at the
impairment date, or at any given cohort date.
Accuracy Ratio (AR)
An accuracy ratio based on the loss experience of structured finance securities (or an accuracy ratio adjusted
for loss-given-default, or LGD) is the ratio of the area between a loss-based cumulative accuracy profile (CAP)
curve and the 45-degree line to the maximum possible area above the 45-degree line. The loss-based CAP
curve (or a CAP curve adjusted for LGD) plots, for each rating category, the proportion of the losses of all
impaired securities accounted for by securities with the same or lower rating against the proportion of all
securities in the sample population with the same or lower ratingS
To calculate accuracy ratios, rating cohorts are formed for each calendar month over the study period so that
all outstanding securities in the sample population are grouped together by their alpha-numeric ratings at the
beginning of the month. For each monthly rating cohort, we determine the number of securities that became
impaired within one or five years of the cohort formation date and their loss severity rates as a percentage of
the cohort date balance. Cumulative shares of securities rank-ordered by rating are calculated for the
universe of all securities and the universe of impaired securities, respectively, and based on this, a CAP curve
is plotted. Note that only LGD for principal impaired securities are used in the calculation. Please see the
definition of LGD for further details.
The CAP curve adjusted for LGD is also known as a "power curve" because it shows how effective a rating
system is at differentiating between securities that have sustained high losses from securities that have
sustained low or no losses. The metric is defined relative to the distribution of ratings in the population. The
accuracy ratio is an effective way to summarize the CAP curve into a single number. The accuracy ratio is
zero if the CAP curve collapses to the 45-degree line, suggesting that all impaired securities are randomly
distributed throughout the population without regard to rating.
8
For an illustration of the CAP curve adjusted for LGD, see "Default & Loss Rates of Structured Finance Securities: 1993-2003," Moody's Speciaf
Comment, September 2004.
iiiiiii:;:iii:iiiii{July 2008 • Special Comment • Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Investment-Grade Loss Rate
The one-year investment-grade loss rate is calculated as follows. First, for a given cohort, we calculate the
LGD as a share of the tranche balance at the cohort date for each security that carried an investment-grade
rating at the cohort formation date and became impaired within a 12-month period after the cohort date. We
then take the sum of these LGD rates and divide by the total number of investment-grade securities
outstanding as of the cohort formation date. Note that the LGD rate is not weighted by the total dollar volume
of outstanding securities in the cohort. Also note that only LGD for principal impaired securities are used in the
calculation. The five-year investment-grade loss rate is calculated similarly.
Average Rating Before Impairment
The rating of an impaired security is measured every month for 36 months prior to impairment. These 36 rating
measurements are averaged together to create one representative number for each impaired security. For a
particular cohort, the average rating before impairment is the weighted average of these average ratings for
each security that became impaired within 12 months after the cohort formation date, where the weight for
each security is the LGD rate of the tranche as a share of its original balance. This weighting scheme will
place greater emphasis on the average ratings of impaired tranches with higher LGD over impaired tranches
B
with lower LGD Note that only LGD for principal impaired securities are used in the calculation.
Rating Action Rate
The rating action rate is defined as the number of securities that experienced a rating change within a year
after cohort formation divided by the total number of securities outstanding at the cohort formation date (the
beginning of each month). Rating changes are measured on the alpha-numeric (or modified) rating scale and
are based on comparing the rating at the beginning and end of the time period under consideration. However,
if a rating was withdrawn by the end of the time period, then the rating prior to withdrawal is used as the end
rating. Note that a security will only be counted if it was outstanding as of the cohort formation date.
Downgrade rates and upgrade rates are measured similarly based on downgrade rating actions and upgrade
rating actions, respectively.
Large Rating Action Rate
A large ration action is said to occur if a rating action (or cumulative rating actions) cause(s) a security's rating
to change by three or more notches within a year after cohort formation. The large rating action rate is the
number of such securities divided by the total number of securities outstanding at the cohort formation date.
Large downgrade rates and large upgrade rates are measured similarly based on large downgrade rating
actions and large upgrade rating actions, respectively.
Percentage of Downgrades (Upgrades) Preceded by Watchlist Actions
in the Same Direction
This metric is defined as the total number of downgraded (upgraded) securities that were placed on the
watchlist in the same direction before they were downgraded (upgraded), divided by the total number of
securities that were downgraded (upgraded) within 12 months after the cohort formation date.
ABS ex HEL
ABS stands for asset-backed securities. This structured finance sector includes securities backed by asset
types such as auto loans, credit card receivables, student loans, and manufactured housing loans, and nontraditional asset types such as mutual fund fees, tax liens, tobacco settlement payments, and intellectual
property. Home equity loans (HEL) are excluded from this sector.
We began using LGD rates as weights in computing an average rating before impairment in the full-year 2005 structured finance performance report.:
Ideally, LGD rates should be calculated as a percentage of the principal balance outstanding for each month in the 36 months prior to the impairment,!
and ratings should then be weighted by these monthly LGD rates. Practically, however, it does not make any material difference in the average rating!
number whether we use LGD rates as a share of impairment-date balance, original balance, or monthly principal outstanding. Since the monthly LGD:
rates are very time consuming to compute due to amortization, we use the LGD rate as a share of original balance as the weight variable.
.
9
iiiiiii:;:i(:i::iii{July 2008 • Special Comment • Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
HEL
The home equity loan or HEL sector includes securities backed by subprime (B&C) mortgage loans, home
improvement loans, high loan-to-value (high LTV) loans, home equity lines of credit (HELOCs), and closedend second-lien loans, as well as net interest margin (NIM) securitizations. It does not include securities
backed by Alt-A mortgages, which are included in the RMBS sector.
Prior to 1998, RMBS collateral was generally defined as first-lien residential mortgages, regardless of the
credit quality of the borrower. HEL collateral generally included junior liens such as HELOCs or closed-end
seconds. However, as subprime lending became more prevalent, the market shifted its definition such that
HEL encompassed subprime first-lien residential mortgages while RMBS included first-lien mortgages made to
higher quality borrowers. Since 1998, a deal classified as RMBS by Moody's is generally backed by prime or
Alt-A quality first-lien residential mortgages, while a deal classified as HEL is generally backed by subprime
first-lien mortgages or junior liens. Therefore, a subprime deal which would be classified as HEL today may
have been classified as RMBS in the past.
RMBS
RMBS stands for residential mortgage-backed securities. The vast majority of these securities are backed by
first-lien prime mortgages or by Alt-A mortgages.
CMBS
CMBS stands for commercial mortgage-backed securities. Commercial real estate (CRE) COOs, where 70%
or more of the collateral is comprised of CRE loans, are classified as CMBS. If the collateral backing the
transaction contains less than 70% CRE loans, then the deal is classified as a COO.
COOs
COOs stand for collateralized debt obligations. Derivative securities such as structured notes and repackaged
securities are not considered to be part of this sector. Commercial real estate (CRE) COOs, where 70% or
more of the collateral is comprised of CRE loans, are classified as CMBS. If the collateral backing the
transaction contains less than 70% CRE loans, then the deal is classified as a COO.
Other Structured Finance
Other structured finance consists of structured finance securities not categorized in the five major sectors
(ABS, HEL, COO, CMBS, and RMBS) including asset-backed commercial paper (ABCP) programs, structured
investment vehicles (SIVs), structured covered bonds, insurance-linked securities such as catastrophe bonds,
and derivative product companies. However, notes carrying only short-term ratings such as commercial paper
are excluded.
Global Structured Finance
Global structured finance captures securities issued around the world in the five major sectors - ABS, HEL,
COO, CMBS, and RMBS - and in the other structured finance category.
u.S. Structured Finance
u.S. structured finance securities are denominated in U.S. dollars and issued in the U.S. market or
denominated in Canadian dollars and issued in Canada. In cases where the source of the underlying collateral
and the denomination of the securities crossed multiple countries/regions, deals are classified by the location
at which they are monitored.
Inti SF ex COO and Other SF
This refers to securities that are not denominated in U.S. dollars and issued in the U.S. market and not
denominated in Canadian dollars and issued in Canada. The majority of the securities in this sector are issued
in Europe, the Middle East, and Africa (EMEA); the rest are issued in the Asia Pacific region and Latin
America. COOs and Other SF are excluded.
iiiiiiii;:iii:;:;iii{July 2008 • Special Comment • Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Related Research
•
The Performance of Structured Finance Ratings: Mid-Year 2007 Report, October 2007 (105390)
•
The Performance of Structured Finance Ratings: Full-Year 2006 Report, May 2007 (103017)
•
The Performance of Structured Finance Ratings: Mid-Year 2006 Report, September 2006 (99034)
•
The Performance of Structured Finance Ratings: Full-Year 2005 Report, May 2006 (97346)
•
The Performance of Structured Finance Ratings: Mid-Year 2005 Report, September 2005 (94463)
•
Default & Loss Rates of Structured Finance Securities: 1993-2007, July 2008 (109707)
•
Measuring Loss-Given-Default for Structured Finance Securities: An Update, December 2006 (101284)
•
Structured Finance Rating Transitions: 1983-2007, February 2008 (107444)
•
Japanese Structured Finance Rating Transitions: 1994-2007, March 2008 (107833/ SF125229)
•
Asia-Pacific (ex-Japan) Structured Finance Rating Transitions: 1990-2007, March 2008 (107947)
•
EMEA Structured Finance Rating Transitions: 1988-2007, March 2008 (107977)
•
Deal Sponsor and Credit Risk of U.S. ABS and MBS Securities, December 2006 (100872)
•
The Relationship between Par Coupon Spreads and Credit Ratings in US Structured Finance, December
2005(95494)
•
Structured Finance Watch list Resolutions: 1992-2003, June 2004 (87305)
•
The Performance of Moody's Corporate Bond Ratings: March 2008 Quarterly Update, May 2008 (108855)
•
Measuring the Performance of Corporate Bond Ratings, April 2003 (77916)
•
Guide to Moody's Default Research: March 2008 Update, March 2008 (108157)
To access any of these reports, click on the entry above. Note that these references are current as of the date of publication
of this report and that more recent reports may be available. All research may not be available to all clients.
iiiilI:i:;iii{July 2008 • Special Comment • Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Report Number:
Julia Tung
Geraldine Kim
© Copyright 2008, Moody's Investors Service, Inc. and/or its licensors and affiliates including Moody's Assurance Company, Inc. (together, "MOODY'S"). All rights
reserved. ALL INFORMATION CONTAINED HEREIN IS PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR
OTHERWISE REPRODUCED, REPACKAGED, FURTHER TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED
FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN WHOLE OR IN PART, IN ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER, BY ANY
PERSON WITHOUT MOODY'S PRIOR WRITTEN CONSENT. All information contained herein is obtained by MOODY'S from sources believed by it to be accurate
and reliable. Because of the possibility of human or mechanical error as well as other factors, however, such information is provided "as is" without warranty of any
kind and MOODY'S, in particular, makes no representation or warranty, express or implied, as to the accuracy, timeliness, completeness, merchantability or fitness
for any particular purpose of any such information. Under no circumstances shall MOODY'S have any liability to any person or entity for (a) any loss or damage in
whole or in part caused by, resulting from, or relating to, any error (negligent or otherwise) or other circumstance or contingency within or outside the control of
MOODY'S or any of its directors, officers, employees or agents in connection with the procurement, collection, compilation, analysis, interpretation, communication,
publication or delivery of any such information, or (b) any direct, indirect, special, consequential, compensatory or incidental damages whatsoever (including without
limitation, lost profits), even if MOODY'S is advised in advance of the possibility of such damages, resulting from the use of or inability to use, any such information.
The credit ratings and financial reporting analysis observations, if any, constituting part of the information contained herein are, and must be construed solely as,
statements of opinion and not statements of fact or recommendations to purchase, sell or hold any securities. NO WARRANTY, EXPRESS OR IMPLIED, AS TO
THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR
OTHER OPINION OR INFORMATION IS GIVEN OR MADE BY MOODY'S IN ANY FORM OR MANNER WHATSOEVER. Each rating or other opinion must be
weighed solely as one factor in any investment decision made by or on behalf of any user of the information contained herein, and each such user must accordingly
make its own study and evaluation of each security and of each issuer and guarantor of, and each provider of credit support for, each security that it may consider
purchasing, holding or selling.
MOODY'S hereby discloses that most issuers of debt securities (including corporate and municipal bonds, debentures, notes and commercial paper) and
preferred stock rated by MOODY'S have, prior to assignment of any rating, agreed to pay to MOODY'S for appraisal and rating services rendered by it fees ranging
from $1,500 to approximately $2,400,000. Moody's Corporation (MCa) and its wholly-owned credit rating agency subsidiary, Moody's Investors Service (MIS), also
maintain policies and procedures to address the independence of MIS's ratings and rating processes. Information regarding certain affiliations that may exist
between directors of MCa and rated entities, and between entities who hold ratings from MIS and have also publicly reported to the SEC an ownership interest in
MCa of more than 5%, is posted annually on Moody's website at www.moodys.com under the heading "Shareholder Relations - Corporate Governance - Director
and Shareholder Affiliation Policy."
Moody's Investors Service
::::lI:;:::::{ July 2008
•
Special Comment •
Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2007 Report
Special Comment
September 2005
Contact
Phone
New York
Jian Hu
Richard Cantor
1.212.553.1653
The Performance of Structured Finance Ratings:
Mid-Year 2005 Report
Highlights
This Special Comment updates Moody's structured finance rating performance metrics, both with respect to rating
accuracy and rating stability.
As indicated in Figure 1, the performance of the most recent rating cohort was similar to the historical average for
previous cohorts across a number of metrics; however, recent performance has been weaker with respect to the 5-year
horizon accuracy ratio (AR). Please see the Glossary at the end of this report for a definition of terms used. Highlights
of this report include:
• The most recent cohort's I-year AR is 74.3 %, up sharply from its level six months ago and roughly equal to
its long-run historical average.
• The 5-year horizon AR for the most recent cohort is 66.3 %, below both its level six months ago and its historical average.
• The I-year investment-grade loss rate is 0.26%, down substantially from its levels six months ago and
slightly below its historical average.
The
36-month average rating of all securities that become impaired over a 12-month period is BaI, the
•
same as its recent past and its long-term historical average.
• The rating action rate is 9.3%, higher than both its levels six and twelve months ago and its historical average, as a sharp increase in the upgrade rate more than offset a concurrent decline in the downgrade rate.
• The large rating action rate (rating changes of three notches or more) is 4.3 %, slightly lower than the rates
observed in the recent past but still above its historical average.
Figure 1 - Summary of All Structured Finance Rating Performance
Accuracy Measures
Cohort Ending Date
June 2005
December 2004
June 2004
December 2003
Average (1993-Most Recent)
Number of
Newly
Impaired
Tranches
over Prior
12 Months
NA
236
252
260
88
Moody's Investors Service
Global Credit Research
1-Year
Accuracy
Ratio
NA
74.3%
68.9%
75.4%
74.7%
5-Year
Accuracy
Ratio
NA
66.3%
69.1%
69.1%
71.5%
1-Year
InvestmentGrade loss
Rate
NA
0.26%
0.41%
0.44%
0.35%
Stability Measures
36-MonthAverage
Rating
Before
Impairment
NA
Bal
Bal
Bal
Bal
1-Year
Rating
Action Rate
9.3%
8.8%
8.8%
10.6%
7.7%
1-Year
large
Rating
Action Rate
4.3%
4.4%
4.6%
6.5%
3.9%
Across the major asset classes (see Figure 2):
• US ABS one-year rating accuracy ratio was 74.2 % for the most recent cohort, slightly lower than its historical
average of 76.4%. The rating action rate stood at 7.7%, roughly the same as its historical average of7.2%.
• US CMBS one-year rating accuracy ratio for the latest cohort was 87.4%, about six percentage points
higher than its historical average, as all newly impaired US CMBS tranches in 2004 were rated below
investment grade at origination. Meanwhile, an increase in upgrade activity caused the rating action rate to
reach 18.4%, far above its historical average of9.8%.
• US RMBS one-year rating accuracy ratio was 61.7% for the most recent cohort, substantially lower than its
historical average of 81.9%. It should be noted that the most recent cohort had only eight tranches that
were newly impaired, whereas historically the total number of impaired tranches is 141. The latest rating
action rate was 8.5% in this sector, higher than its historical average of 6.2%. The majority of the latest rating actions were upgrades, but the number of downgrades also increased.
• Global CDO one-year rating accuracy ratio was 67.9% for the most recent cohort, slightly below its historical average of 69.5%. The sector's latest rating action rate was 7.5%, substantially lower than its historical
average of 13.6%. Furthermore, for the first time since 1998, the CDO sector saw the upgrade rate surpass
the downgrade rate in the most recent cohort.
• International Structured Finance one-year rating accuracy ratio was 88.0% for the most recent cohort,
which had only two newly impaired securities. Overall, the number of impaired securities in this sector is
too small for any meaningful comparisons over time and across sectors. The latest rating action rate of the
sector stood at 2.9%, which was above its historical average level of 1.8%; and the number of upgrades significantly outweighed the number of downgrades.
Figure 2 - Summary of Structured Finance Rating Performance by Sector
Accuracy Measures
Number of
Newly
Impaired
Tranches over
Prior 12 Months
US ABS
US CMBS
US RMBS
Global COOs
Int'I SF excluding CDOs*
143
42
8
41
2
US ABS
US CMBS
US RMBS
Global COOs
Int'I SF excluding CDOs*
43
7
12
26
0
* International structured finance excluding COOs
2
Moody's Special Comment
36-Month1-Year
Average
1-Year
5-Year
InvestmentRating
Grade loss
Before
Accuracy
Accuracy
Ratio
Ratio
Rate
Impairment
Cohorts Ending December 2004
74.2%
66.7%
0.46%
Baa2
87.4%
87.1%
0.00%
B2
61.7%
87.7%
0.05%
Baa3
67.9%
55.3%
0.46%
Bal
88.0%
95.4%
0.02%
Bal
Historical Averages Since 1993
76.4%
74.9%
0.49%
Baa3
81.6%
86.8%
0.07%
Bl
81.9%
74.7%
0.11%
Bal
69.5%
58.4%
1.07%
Ba2
62.1%
94.4%
0.02%
Baa2
Stability Measures
1-Year
1-Year
large
Rating
Rating
Action
Action
Rate
Rate
Cohorts Ending June 2005
7.7%
4.2%
18.4%
5.2%
8.5%
5.1%
7.5%
3.7%
7.4%
2.9%
Historical Averages Since 1993
7.2%
4.6%
9.8%
3.2%
6.2%
2.9%
13.6%
7.5%
5.2%
1.8%
Table of Contents
Page
Introduction .................................................................................................................................... 4
Accuracy Ratios .............................................................................................................................. 6
Investment-Grade Loss Rates ......................................................................................................... 7
Average Rating Before Impairment ................................................................................................. 8
Rating Action Rates, Large Rating Action Rates, and Watchlist Actions ........................................... 9
Glossary ....................................................................................................................................... 13
Related Research ......................................................................................................................... 15
Moody's Special Comment
3
Introduction
In a Special Comment published in April 2003, Moody's developed a set of metrics to measure the performance of
corporate ratings with respect to the dual objectives of rating accuracy and rating stability. 1 Moody's corporate rating
performance report has since been updated on a ~uarterly basis. 2 Although these metrics have also been reported for
structured finance ratings in various prior studies, this Special Comment groups them together for the first time in a
single document.
This report is similar in design in many respects to the corporate rating performance report. The basic unit of
observation is a monthly cohort of ratings, i.e. all outstanding ratings at the beginning of a month, and the metrics
compare the performance of different rating cohorts over time. Ratings on individual tranches are all given equal
weight in the analysis, independent of the tranche size. 4 This report differs, however, from the corporate report in
that:
• The structured finance metrics on rating accuracy lag those for corporate finance by six months due to the
need to identify defaults and estimate losses from periodic payment reports.
• Structured finance rating accuracy metrics are based on structured securities' impairment experiences and
their estimated final loss severity rates, because Moody's ratings are intended to rank order expected loss
rates, not simply expected impairment rates. S
• The performance of structured ratings are not compared to the performance of ratings implied by bond
market credit spreads because secondary market price data are not as commonly available for structured
finance securities as for corporate securities.
• Data are presented on a disaggregated basis primarily by asset class for deals issued in the United States.
International structured finance securities outside of collateralized debt obligations are presented as a separate category.
The most important measure of rating accuracy is the accuracy ratio, which for structured finance measures the
relationship between security ratings and their realized loss rates. 6 This metric measures the quality of Moody's ratings
as indicators of relative expected credit loss risk.
As discussed in the April 2003 Special Comment, although relative rating accuracy is our primary objective,
Moody's recognizes that many investors are also concerned with the cardinal accuracy of the rating system. In particular, they expect that investment-grade credits should rarely suffer credit losses and impaired securities should normally carry low ratings well in advance of impairment. For this purpose, we will regularly track investment-grade loss
rates and the average rating of securities during the 36 months prior to impairment. Both of these measures should be
low if the rating system is accurate in a cardinal sense.
Measures of cardinal rating accuracy can also be found in Moody's default and loss studies for structured finance
securities. In particular, we have compared realized and "idealized" loss rates over various investment horizons by rating category. The idealized loss rates are used as "targets" in the structured finance rating process. Because realized
loss rates are highly volatile, semi-annual comparisons between realized loss rates and idealized rates are not statistically meaningful, and hence not tracked in our rating performance reports.
We employ two measures of rating stability (or rating volatility) - the rating action rate (the frequency of rating
changes) and the large rating action rate (the frequency of rating changes of three notches or more) over a twelvemonth period. To examine how watchlist actions are used, we also report the percentages of downgrades and upgrades
preceded by watchlist (review) actions.
1.
2.
3.
4.
5.
6.
4
"Measuring the Performance of Corporate Bond Ratings," Moody's Special Comment, April 2003.
For the latest performance report, see "The Performance of Moody's Corporate Bond Ratings: June 2005 Quarterly Update," Moody's Special Comment, July 2005.
Moody's first introduced and examined its structured finance ratings' performance metrics in "Default & Loss Rates of Structured Finance Securities: 1993-2003,"
Moody's Special Comment, September 2004.
Rating cohorts consist of all outstanding security ratings, both newly issued and seasoned securities. In our annual rating transition and default and loss studies, we
report selected accuracy and stability measures on both cohort and vintage bases, and on equal-weighted and volume-weighted bases.
As reported in our July 2005 Special Comment entitled, "Default & Loss Rates of Structured Finance Securities: 1993-2004," loss-given-impairment rates vary
sharply by rating category in structured finance, with higher rated issues suffering much lower average loss severity upon impairment than lower rated issues. In contrast, as reported in our January 2005 Special Comment "Default and Recovery Rates of Corporate Bond Issuers: 1920-2004," loss-given-default rates do not vary
strongly by issuer rating categories prior to default in the corporate sector. For this reason, and because we lack loss severity data for many defaulted corporate securities, our reported corporate rating performance statistics are based simply on default rates, rather than loss rates.
The required adjustments to convert the standard default-based AR measure to a loss rate-based measure are discussed in "Default & Loss Rates of Structured
Finance Securities: 1993-2003," Moody's Special Comment, September 2004. The accuracy ratio is also described in the Glossary at the end of this report.
Moody's Special Comment
We calculate accuracy statistics based on the data from Moody's latest structured finance default and loss study
(1993-2004).7 For the measurement of stability and volatility, we incorporate the most recent rating actions observed
in the first half of 2005 into the data sample used in Moody's latest structured finance rating transition study.8 The
data sample of this report includes all public, 144A, and private tranches with a Moody's long-term global debt rating
in the global asset-backed securities, commercial and residential mortgage-backed securities and collateralized debt
obligations sectors. Prospective ratings, credit evaluations, and national scale ratings are not included. Pari passu
tranches are collapsed into a single tranche. In addition, the following tranches are excluded: (1) Tranches guaranteed
by financial guarantors or govemment-sponsored-enterprises (GSEs); (2) Interest-only (10) or residual tranches; (3)
Repackaged securities, structured notes, structured investment vehicles, structured covered bonds, and other credit
derivative securities; (4) Deals that have all their tranche ratings linked to a single corporate or sovereign rating.
7.
8.
"Default & Loss Rates of Structured Finance Securities: 1993-2004," Moody's Special Comment, July 2005.
"Structured Finance Rating Transitions: 1983-2004," Moody's Special Comment, February 2005.
Moody's Special Comment
5
Accuracy Ratios
Figure 3 depicts the one- and five-year accuracy ratios over time:
• The one-year and five-year accuracy ratios in the all structured finance category have generally trended
down over the past ten years, primarily as a result of the severe credit distress in two structured finance segments: ABS securities backed by manufactured housing loans and CDOs backed by high yield corporate
bonds. A substantial portion of the rated tranches in these two segments (some of them highly rated) were
impaired and are expected to sustain high loss-given-default. Figure 3 shows, however, that the one-year
accuracy ratio has showed material improvement in the latest cohorts.
• For most of the study period, the accuracy ratios in the US CMBS and US RMBS were at high levels
between 80% and 90%. The one-year accuracy ratio of US RMBS ratings in the recent cohorts, however,
saw a steep decline primarily because three (out of a total of eight in 2004) newly impaired tranches (securitized by Alt-A mortgages) carried single-A ratings at the time of impairment. 9
• The one-year accuracy ratio of US ABS ratings recently improved slightly, but remained lower than its historical average of76.4%; whereas, the one-year accuracy ratio of Global CDOs has remained flat at a level slightly below 70%.
Figure 3 - One-Year (solid line) and Five-Year (dashed line) Accuracy Ratios (ARs) By Cohort Date
All Structured Finance
USABS
100%
100%
90%
90%
80%
80%
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
..
,
~
..
.:
'.
,',
'.'
.
'.
"
10%
10%
0%
0%
M
7
~
~
~
~
~
~
~
ro
~
~
~
a
a
~
~
~
~
~
~
~
~
~
a
M
7
a
~
~
~
~
~
~
M
~
a
~
a
7
N
US RMBS
US CMBS
100%
100%
~
90%
80%
J
70%
60%
50%
40%
30%
20%
10%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
0%
M
~
C
~
~
~
~
~
~
~
ro
~
~
~
c
c
c
c
c
c
7
a
a
c
a
c
N
a
c
M
a
c
M
7
a
c
~
c
~
c
Global CDOs
100%
a
a
c
7
N
a
c
7
M
a
c
a
c
a
c
International Structured Finance excluding CDOs
100%
90%
90%
80%
80%
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
J
0%
0%
M
~
~
~
~
~
~
~
~
~
~
~
~
~
7
M
~
c
~
~
~
~
~
~
~
ro
c
c
c
c
c
7
~
a
a
c
a
c
N
a
c
M
a
c
7
a
c
Notes:
• At the beginning of each month, securities are grouped together by their alpha-numeric ratings to form rating cohorts. The latest one- and five-year
cohorts are formed on January 1,2004 and January 1,2000.
• The gaps in the US CMBS and International Structured Finance accuracy ratio series were resulted from zero impairments in those periods. In addition,
the accuracy ratios tend not to be meaningful when the number of impaired securities is very small (in single digits). For example, the five-year accuracy
ratio for international structured finance is based on a single impaired security that was rated Ba2 in 1996 and then impaired in 2002. This structured
finance sector had just five impaired tranches, four of which were rated after January 1, 2000. The impairment experiences of these four tranches are
reflected in the one-year accuracy ratios. Finally, only one CMBS security was impaired in 1994, and it was rated Baa2 before impairment.
9.
6
Due to the small number of impaired securities in the latest cohort, the decline in the accuracy ratio does not represent a trend in the performance of RMBS ratings.
Moody's Special Comment
Investment-Grade Loss Rates
Figure 4 shows that the one-year loss rates of investment-grade securities have declined significantly in the latest
cohorts.
• In the all structured finance category, the investment-grade loss rate dropped to 0.26% in the latest annual
cohort from 0.44% a year prior. This rate is below its historical average of 0.35% and much lower than the
peak of 0.75% observed in the January 2002 cohort. lO
• The US ABS and Global CDO sectors saw the most significant improvement, but their investment-grade
loss rates remained higher than those in the US CMBS, US RMBS, and international structured finance
sectors.
Figure 4 - One-Year Investment-Grade loss Rates By Cohort Date
All Structured Finance
USABS
1.4%
1.0%
0.9%
1.2%
0.8%
0.7%
0.6%
1.0%
0.8%
0.5%
0.4%
0.3%
0.2%
0.1%
0.0%
0.6%
0.4%
0.2%
co
m
.,.
m
m
c
c
c
ro
ro
CD
ro
;
c-
m
oo
m
m
m
c
c
c
ro
ro
a
~ ~ ~
~ ~
ro
0.0%
co
.,.
m
m
c
c
c
ro
ro
;
CD
m
ro
m
c-
m
oo
m
m
c
c
c
ro
ro
a
~ ~ ~
~ ~
ro
US RMBS
USCMBS
1.0%
1.8%
1.6%
1.4%
1.2%
1.0%
0.8%
0.6%
0.4%
0.2%
0.0%
0.9%
0.8%
0.7%
0.6%
0.5%
0.4%
0.3%
0.2%
M
'J;
-":
'J;
'J;
U?
<D
'J;
'J;
'J;
'J;
en
co
-":
-":
-":
-":
-":
-":
-":
..".
c-
oo
~
~
-":
N
M
..".
-":
-":
-":
~
~
0.1%
0.0%
~
co
.,.
CD
m
m
m
c
c
c
ro
ro
ro
;
m
c-
m
oo
m
m
c
c
c
ro
ro
a
ro
~ ~ ~
~ ~
International Structured Finance excluding CDOs
Global CDOs
1.0%
3.0%
0,9%
2.5%
0.8%
0.7%
2.0%
0.6%
0.5%
1.5%
0.4%
0.3%
1.0%
0.2%
0.5%
0.1%
0.0%
0.0%
co
.,.
CD
m
m
m
~
~
~
'"m
~
c-
oo
m
m
m
m
~
~
~
a
a
~
a
~
N
a
~
co
a
~
.,.
a
~
co
.,.
CD
m
m
m
c
c
c
ro
ro
ro
'"mc
ro
m
c-
m
oo
m
m
c
c
c
ro
ro
ro
a
a
c
ro
a
c
ro
N
a
c
ro
co
a
c
ro
.,.
a
c
ro
Note: At the beginning of each month. securities carrying an investment-grade rating are grouped together to form a rating cohort. The latest rating
cohort is formed on January 1. 2004.
10. We note that these loss rates are not weighted by dollar volume. In addition. like-rated tranches in the same deal are collapsed into a single tranche. The average rating of investment-grade securities in structured finance has not changed much historically and currently is in between Aa3 and A 1.
Moody's Special Comment
7
Average Rating Before Impairment
Figure 5 presents the 36-month average rating before impairment over time. The figure also contains the number of
impaired securities used to calculate the average ratings over time. The following observations are noteworthy:
•
•
•
•
The average rating before impairment in the all structured finance category has increased in the last four years, from
roughly Ba2 for the cohorts formed in the beginning of 2000, to somewhere between Ba1 and Baa3 for the latest cohorts.!!
The average rating before impairment in US C'MES is the lowest among all sectors; in fact, it is often significantly
lower than those in the other sectors. Furthermore, the sector's average rating before impairment has been improving,
and reached B2 for the most recent cohort.
The average rating before impairment in the Global CDO sector has slightly increased from roughly Ba3 for the
cohorts formed in 1999 to Ba1 for the most recent cohorts.
The average ratings before impairment in the US ABS and RMBS sectors have been volatile historically, especially during periods in which the number of newly impaired securities was small. The average rating before impairment in the
US ABS sector has trended up and approached Baa2 in the most recent cohorts, as a result of more investment-grade
tranches backed by manufactured housing (MH) loans being impaired in 2004. For the RMBS sector, the average rating also trended up in the latest cohort. Of the eight impaired tranches, five were rated investment-grade, and that was
only 0.15% of roughly 3,500 investment-grade RMBS tranches in the cohort.
Figure 5 - 36-Month-Average Ratings Before Impairment (solid line, left axis) and Number of Newly
Impaired Securities (dashed line, right axis) By Cohort Date
USABS
All Structured Finance
450
400
Baa2
Baa3
350
300
Ba1
Ba2
250
200
Ba3
B1
150
100
B2
50
B3
~
~ ~ ~ ~ ~ ~ ~
200
Baa3
Ba1
150
Ba2
100
Ba3
B1
50
B2
B3
a
~
~
~
~
~
~
~
~
~ ~ ~
a
us RMBS
us CMBS
Baa2
Baa3
r
01:
80
Ba1
Ba2
Ba3
B1
B2
B3
Caa1
"1···~··l······r·····'······r·····'······r···
a
~
Baa2
11111
Baa1
Caa2
250
Baa1
500
Baa1
100
Baa1
Baa2
80
Baa3
Ba1
60
Ba2
40
Ba3
B1
20
B2
'
B3
0
....
~
i i i i i i i ~ 1~ ~ ~
i i i i i i i 11111
Intemational Structured Finance excluding CD Os
Global COOs
250
Baa1
Baa2
10
A1
Baa1
200
Baa3
Baa2
Baa3
Ba1
150
Ba1
Ba2
100
Ba3
Ba2
Ba3
B1
.,-,'
50
B2
B1
B2
B3
B3
~
~
~
~
~
~
c
-'i
~
~
~
~
~
i i i i i i i ~ 1 ~ ~ ~
~
Notes:
• At the beginning of each month. all securities that become impaired within the next 12-month period are grouped to form a cohort. The latest cohort is
formed on January 1. 2004.
• The number of newly impaired securities in the latest annual cohort is eight in the US RMBS sector, and two in the International Structured Finance
sector. The international structured finance sector had three impaired securities in 2001. all of which were from deals issued in Latin America.
11. The average rating before impairment of Ba 1 in the structured finance has historically averaged four notches higher than the comparable statistic for corporate finance
sector of B2. While this metric indicates lower rating accuracy for structured finance. other accuracy metrics (investment-grade loss rates and accuracy ratios) suggest the relative rating accuracy of the two sectors is closer. Moreover, the validity of rating accuracy comparisons across sectors (as opposed to comparisons over
time within sector) is unclear, particularly in light of the very different overall rating distributions of the two sectors. In particular, the average rating in structured finance
of A 1 is also four notches higher than the average rating in corporate finance of Baa2.
8
Moody's Special Comment
Rating Action Rates Large Rating Action Rates, and Watchlist Actions
Figure 6 reports rating action rates and large rating action rates. Figures 7 and 8 further disaggregate rating actions
into downgrades and upgrades, and Figure 9 demonstrates how frequently downgrades and upgrades have been preceded by watchlist actions. 12 Key observations include:
•
•
•
•
Rating action rates in the all structured finance category increased in the first half of 2005, as a sharp
increase in upgrades in the US CMBS and US RMBS sectors more than offset a concurrent decline in
downgrades in US ABS and global CDO sectors. As a result, the upgrade-to-downgrade ratio rose to 2.5to-l in the first half of 2005, a significant improvement over the ratio of I-to-l in 2004.
The downgrade rate has declined substantially to 3.1 % in the latest one-year cohort ending June 2005,
from 5.8% a year prior. About 92 % of the downgrades in the first half of 2005 were the result of weaker
collateral performance, roughly the same as that in 2004.
The upgrade rate is twice as high, at 6.2%, in the latest annual cohort, as its level a year earlier. Most of the
upgrades were observed in US CMBS and US RMBS sectors. About 93 % of upgrades stemmed from a build
up in credit enhancement thanks to deal seasoning andlor stronger collateral performance.
Downgrades have frequently been anticipated by rating reviews, especially in recent annual cohorts, for
which 80% of all downgrades were preceded by watchlist actions. Upgrades, however, have been less commonly preceded by watchlist actions, although the percentage of upgrades anticipated by review has
increased in the recent cohorts.
Figure 6 - Annual Rating Action Rates (dashed line) and large (Three Notches or More) Rating Action
Rates (solid line) By Cohort Date
All Structured Finance
USABS
15%
15%
12%
12%
"
....
9%
6%
3%
9%
""':": .. " " . '
~
••
.,.",
,:"
,,:,;
",
.....
..
,:
6%
:"
";
3%
O%~~~~~~--~~~~~~~
~
~
~
~
~
~
~
~
~
~
~
~
0%
~
8
qt;
~
~
~ ~
S=;
~
US CMBS
12%
0%
j
~
8
~
~
l
j
20%
ti.·······.· · · . · ·
16%
12%
i . . .': :'.' ....:
40"f:.~
'1·
...
8%
1]
~
US RMBS
20%
16%
~
I'"
.'.
8%~
. . . : . . . . . . . . . . . . . . . . . . . . . ... .
.'
4%
~
". ,:' ,"
'.'
~
. ",
.:"
..
.
+-~~~~"'----~~~~---;=.,--~
~ ~ ~
" ~
~
""
':.,~,'
.
0%
~ ~ ~ ~ ~ ~ ~ ~ ~ "
'.
Global CDOs
~
'"
~
~
i
~ ~ ~ ~ ~ ~
International Structured Finance excluding COOs
30%
30%
25%
25%
20%
~
20%
..
15%
"
.
",:
"
10%
,.'
,
::IJ~~;~~~~.t~.~.~--·~·-·~·-··~~~~·-·~
B:
~
~
B:
~
~
B:
15%
10%
5%
I
.. ~.~
0%10
::
~
12. For the latest rating transition statistics of structured finance securities worldwide, see "Structured Finance Rating Transitions: 1983-2004," Moody's Special Comment, February 2005. Moody's also reviewed its uses of watchlist actions on structured finance securities in a Special Comment, "Structured Finance Watch list Resolutions: 1992-2003," June 2004.
Moody's Special Comment
9
Figure 7 - Annual Downgrade Rates (dashed line) and Annual large (Three Notches or More)
Downgrade Rates (solid line) By Cohort Date
US ABS
All Structured Finance
US RMBS
US CMBS
8%
6%
Global COOs
International Structured Finance excluding COOs
15%
12%
9%
6%
:::b
I·
10
Moody's Special Comment
Figure 8 - Annual Upgrade Rates (dashed line) and Annual large (Three Notches or More) Upgrade
Rates (solid line) By Cohort Date
All Structured Finance
US ADS
10%
10%
8%
8%
-,:
6%
6%
4%
4%
2%
0%
~
i
':':.
M
~
~ ~ ~
'n
Q
~
~
~
2%
. ::'., ...,"
m
~
0
~ ~ ~
T
~
~.
0%
t~~
~
M
~
~
~:
~
~ ~ ~ '"
~ ~ ~
0,
~
'"
12%
8%
6%
4%
3%
2%
0%
0%
'-
~
0'
."
~
~,
-~
co
0'
"
~,
-~
~
~
0'
0:
"
".
C~
a
~,
~
~
~ ~ ~ ~
6%
,
::"
~
a
'"
US RMBS
10%
"-
Q"l
'"
US CMBS
15%
9%
~
'" ~ '"
0
0
"
~
-~
~
N
0
c:
".
"
~
0;
c:
".
"
~
~
M
m
~
~
m
~
~
~
~
~
~
~
Global COOs
,.m
~
00
~
~,
Q
~
m
0
0
0
~
~
~
~
~
"0
~
-,
c
,n
~
~
0
International Structured Finance excluding COOs
15%
8%
12%
6%
9%
4%
6%
2%
3%
0%
0%
(J'
0'
"'
-',
"
--,
M
7
(J'
~
0'
"
~
"-
'-
ro
0'
0'
~
."
"
0
c:
--.
7
N
0
"
c:
--.
~
0
~
a
"
--,
"
M
0'
~
0)
'n
~
co
e,
~
a,
~
~
~
m
e,
a
0
a
N
8
~
~
~
0
0
0
~
Moody's Special Comment
11
Figure 9 - Percentages of Downgrades (solid line) and Upgrades (dashed line) Preceded by
Watchlist Actions By Cohort Date
All Structured Finance
US ABS
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
n
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
<C,
'"
C
~
"
~
Gl
~:
en
C
'"
0:
~
~
0'
C
'"
~
Gl
0:
en
,~
<0
::,
C)
<0
"
~
c
c
~
'"c
00
~
"c:
<0
~
~
'--]"-'
m
'"
c
"'c
en
m
en
""
~
C
N
'0
<D
c-
m
en
C
C
N
C
N
(~..i
..
: ":
~l
:':
eo
m
c
~
".
m
en
c
c
~
,n
~
m
"'
~
"
,-.
m
c
~
oo
m
c
~
en
en
c
~
co
'"c
~
co
c
~
co
'"
c
~
,>,
co
~
",.
'"c
~
en
co
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
c
C
N
m
m
co
'D
cen
oo
m
en
en
~
C
C
~
~
C
(~..i
C
N
.".
m
""
C
co
C
'0
".
co
eo
.".
~
C
'"
C
N
'"
(~..i
en
co
C
N
IT
'0
'"c
~
~
.'
;.
.
,
I;
•.
",.
C~l
"-,
<D
:))
C~l
c
c
c
~
~
~
,-
'"c
~
en
'"
c
m
'"c
~
co
00
c
~
'0
00
N
~~ ~
""
"-,
~ ~~
c
~
International Structured Finance excluding COOs
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
'~"I'-'-i'-
en
'"
~
.,
Global COOs
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
co
m
m
US RMBS
US CMBS
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
oo
m
co
'"
C
'0
N
'"
C
N
co
~
co
'"
C
'0
""
00
C
~
en
c
~
','" :
m
'"
C
~
"
m
~
co
ill
en
m
en
oo
en
en
C
C
C
C
C
~
~
~
m
~
c-
co
'"
C
~
co
N
00
C
C
~
co
'"
C
~
"C
~
co
<0
C
Note: At the beginning of each month, all securities that experienced downgrades (upgrades) within the next 12-month period are grouped together to
form a rating cohort. The latest cohort is formed on July 1, 2004.
12
Moody's Special Comment
MATERIAllMPAIRMENT
Material impairment, or for simplicity, impairment, is a concept adopted in Moody's structured finance default and
loss study. Material impairment includes uncured payment defaults and securities downgraded to Ca or C, where payment defaults include shortfalls of interest and losses of principal on individual tranches of structured finance transactions. A security is called "newly impaired" in a given period if it had no outstanding interest shortfalls, no principal
losses, and was not rated Ca or C in the prior period, but experienced at least one of these three credit events in the
given period for the first time.
lOSS-GIVEN-DEFAUlT (lGD)
Loss-given-default, also known as LGD or loss severity rate, is the total amount of lifetime losses as a share of a
tranche's principal balance on a certain reference date. The losses in each payment period are discounted by a discount
rate, which is typically the stated coupon rate on the tranche. There are three types of principal balances used in the
calculation of LGD: the principal balance at origination, at the time of impairment, and at a cohort formation date.
Depending on the reference principal balance (and the reference date), the calculated LGD can be significantly different due to principal amortization and discounting.
Final LGD on impaired securities with no outstanding principal balances are typically known. For impaired securities with positive principal balances or impaired securities with incomplete loss data, their LGD need to be estimated. Detailed discussions on the estimation of those final LGD are provided in several Moody's Special Comments
including, in particular, "Measuring Loss Severity Rates of Defaulted Residential Mortgage Backed Securities,"
Moody's Special Comment, April 2004.
ACCURACY RATIO (AR)
An accuracy ratio based on structured securities' loss experiences (or an accuracy ratio adjusted for loss-given-default,
or LGD) is the ratio of the area between a loss-based cumulative accuracy profile (CAP) curve and the 45 -degree line to
the maximum possible area above the 45-degree line. The loss-based CAP curve (or a CAP curve adjusted for LGD)
plots, for each rating category, the proportion of the losses of all impaired securities accounted for by securities with the
same or lower rating against the proportion of all securities in the sample population with the same or lower rating. 13
To calculate accuracy ratios, we use the same data sample that is used in the latest structured finance default and
loss study (1993-2004), and form rating cohorts of different horizons for each calendar month during 1993-2004, i.e.
all outstanding securities in the sample population are grouped together by their alpha-numeric ratings at the beginning of the month. We then do the same (group by ratings in each calendar month) for securities that became impaired
within one, three or five years from that month. Cumulative shares of securities rank-ordered by rating are calculated
for the universe of all securities and the universe of impaired securities, respectively, and based on this, a CAP curve is
plotted.
The CAP curve-adjusted for LGD is also known as a "power curve" because it shows how effective a rating system
is at differentiating between securities that have sustained high losses from securities that have sustained low losses or
no losses. The metric is defined relative to the distribution of ratings in the population. The accuracy ratio is an effective way to summarize the CAP curve into a single number. The accuracy ratio is zero if the CAP curve collapses to the
45-degree line, suggesting that all impaired securities are randomly distributed throughout the population without
regard to ratings.
Because we use only data observations that span the time horizon under consideration, the latest cohort for a oneyear horizon is formed onJanuary 1, 2004, while for a five-year horizon the latest cohort is formed onJanuary 1, 2000.
INVESTMENT -GRADE lOSS RATE
The one-year investment-grade (Aaa through Baa3) loss rate is calculated as follows. First, for each impaired security,
we calculate a loss rate, given impairment, as a share of its tranche balance at the cohort date. Then, we sum up these
loss rates to get a total loss value for all securities that become impaired within a 12-month period after the cohort formation date. We divide this total loss value by the total number of securities in the rating cohort to obtain the investment-grade loss rate. Note that this loss rate is not weighted by the total dollar volume of outstanding securities in the
cohort.
13. For an illustration of the CAP curve adjusted for LGD. see "Defauff & Loss Rates of Structured Finance Securities: 1993-2003." Moody's Special Comment. September 2004.
Moody's Special Comment
13
AVERAGE RATING BEFORE IMPAIRMENT
The rating of an impaired security is measured every month for 36 months prior to impairment as well as immediately
prior to impairment. These 37 rating measurements are averaged together to create one representative number for
each impaired security that becomes impaired within 12 months after the month of cohort formation. These representative numbers are then averaged together to create the reported average rating each month.
RATING ACTION RATE
The rating action rate is defined as the number of securities that experienced a rating change within a year after cohort
formation divided by the total number of securities outstanding at the cohort formation date (the beginning of each
month). Rating changes are measured on the alpha-numeric (or modified) rating scales. Downgrade rates and upgrade
rates are measured similarly based on downgrade rating actions and upgrade rating actions, respectively.
LARGE RATING ACTION RATE
A security that had its rating changed by three notches or more within a year after cohort formation is said to have
experienced a large rating action. The large rating action rate is the number of such securities divided by the total
number of securities outstanding at the cohort formation date. Large downgrade rates and large upgrade rates are
measured similarly based on large downgrade rating actions and large upgrade rating actions, respectively.
PERCENTAGE OF DOWNGRADES (UPGRADES) PRECEDED BY WATCH LIST ACTIONS
This metric is defined as the number of securities that were placed on watchlist before they were downgraded
(upgraded), divided by the total number of securities that were downgraded (upgraded) within 12 months after the
cohort formation date.
ABS
This refers to Asset-Backed-Securities. This structured finance sector includes securities backed by home equity loans
(HEL) in addition to both traditional (autos, credit cards, leases, manufactured housing, student loans, etc.) and nontraditional (mutual fund fees, tax liens, tobacco settlement, whole business securitizations (WBS), etc.) asset classes.
HELS
HELs include securities collateralized by subprime (B&C) mortgage loans, home improvement loans, high loan-to-value
(high LTV) loans, home equity lines of credit (HELOCs), closed-end second-lien loans, and net interest margin (NIM)
securitizations. It does not include "Alt-A" mortgages, which are part of the RMBS sector. HEL is part of the ABS sector.
CDOS
This refers to collateralized debt obligations. Credit derivative securities such as repackaged securities and structured
notes are not considered to be part of this sector.
CMBS
This refers to commercial mortgage-backed securities.
RMBS
This refers to residential mortgage-backed securities. The large majority of these securities are backed by first-lien
prime mortgages, although some are backed by Alt-A mortgages. In some older vintage RMBS transactions, subprime
mortgages are also included in the collateral. HEL is not part of this sector.
U.S. STRUCTURED FINANCE SECURITIES
This refers to structured finance securities denominated in U.S. dollars and issued in the U.S. market.
INTERNATIONAL STRUCTURED FINANCE EXCLUDING CDOS
This refers to non-CDO structured finance securities that are not denominated in U.S. dollars or not issued in the
U.S. market. The majority of the securities in this sector are in Europe; the rest comes from the Asia Pacific region,
Canada, and Latin America.
14
Moody's Special Comment
Related Research
Special Comments:
P~f~~!t~__1Q??__R~J~~u?f.~.m~~~x.~g)<\g!~D.~~ __~~.~llX.hi~§:J.?'9.J:.?Q.Q:hJ~Jy}Q.Q.H9.}§U)
Default & Loss Rates of US CDOs: 1993-2003, March 2005 (91692)
Default & Loss Rates of Structured Finance Securities: 2004 First Half Update, January 2005 (90843)
Default & Loss Rates of Structured Finance Securities: 1993-2003, September 2004 (88692)
Measuring Loss Severity Rates of Defaulted Residential Mortgage Backed Securities, April 2004 (86769)
Payment Defaults and Material Impairments of US Structured Finance Securities: 1993-2002,
December 2003 (80247)
~.~fn!r~Q.!:Ig.~!l~~__R!!!!DgT~.~!l?hjQ!l?:.J.9..~J.:.?QQ.:1:,..f~Q~~D!.IQQ.~__~n~~2
.~~~~!~.g..f!!l!!Df.~.R~!:!DgT!!!D?!!!gD?:J~~I:~.Q.QJ.LI~!?~!!Xy.~.Q.Q:H~J.n~)
~.~fn!r~Q.E!D.~!l~~__R!!!!DgT~.~!l?!!:!Q!l?:.J.2.~J.:.~QQL.~g!!!l?~E!?'QD.?__!0.th__~g!l?QE~t.~.R~!iDg?__~Dg__AfXg§§__~~.~tg!§L
I~D~~ry__~QQL(ZZ2:9..!2
.~~~~!.~.g..f!D!!Df.~.W.~.~.c;hI!?tR~?'QJ~!iQD.?:J.?n.:.?QQ},J~~JQQ.1..c~z}Q~2
Q~!.M.~g.MgQgi§__P~f~~!!.R~?~.~!.~hJ~~}Q.Q.?__ yp'.Q!!!~__(~J.QZ~2
The Performance of Moody's Corporate Bond Ratings: June 2005 Quarterly Update, July 2005 (93560)
Measuring the Performance of Corporate Bond Ratings, April 2003 (77916)
To access any of these repo1'ts, click on the entry above. Note that these references are CU17Cnt as of the date ofpublication of this
repo1't and that more recent repo1'ts may be available. All research may not be available to all clients.
Moody's Special Comment
15
To order reprints of this report (100 copies minimum), please call 1.212.553.1658.
Report Number: 94463
Author
Senior Production Associate
Jian Hu
Charles Ornegri
© Copyright 2005, Moody's Investors Service, Inc, and/or Its licensors including Moody's Assurance Company, Inc, (together, "MOODY'S"), All rights reserved, ALL INFORMATION
CONTAINED HEREIN IS PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR OTHERWISE REPRODUCED, REPACKAGED, FURTHER
TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN WHOLE OR IN PART. IN ANY
FORM OR MANNER OR BY ANY MEANS WHATSOEVER, BY ANY PERSON WITHOUT MOODY'S PRIOR WRITTEN CONSENT. All information contained herein IS obtained by
MOODY'S from sources believed by It to be accurate and reliable, Because of the possibility of human or mechanical error as well as other factors, however, such information IS provided "as
IS" without warranty of any kind and MOODY'S, In particular, makes no representation or warranty, express or Implied, as to the accuracy, timeliness, completeness, merchantability or fitness
for any particular purpose of any such Information, Under no circumstances shall MOODY'S have any liability to any person or entity for (a) any loss or damage In whole or In part caused by,
resulting from, or relating to, any error (negligent or otherwise) or other circumstance or contingency within or outside the control of MOODY'S or any of Its directors, officers, employees or
agents In connection with the procurement, collection, compilation, analysIs, Interpretation, communication, publication or delivery of any such information, or (b) any direct, indirect,
special, consequential, compensatory or incidental damages whatsoever (Including without limitation, lost profits), even If MOODY'S IS advised In advance of the possibility of such
damages, resulting from the use of or inability to use, any such information, The credit ratings and financial reporting analysIs observations, If any, constituting part of the information
contained herein are, and must be construed solely as, statements of opinion and not statements of fact or recommendations to purchase, sell or hold any SeCUrities, NO WARRANTY,
EXPRESS OR IMPLIED, AS TO THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR OTHER
OPINION OR INFORMATION IS GIVEN OR MADE BY MOODY'S IN ANY FORM OR MANNER WHATSOEVER. Each rating or other opinion must be weighed solely as one factor In any
Investment decIsion made by or on behalf of any user of the information contained herein, and each such user must accordingly make ItS own study and evaluation of each security and of
each Issuer and guarantor of, and each provider of credit support for, each security that It may consider purchasing, holding or seiling
MOODY'S hereby discloses that most Issuers of debt seCUrities (including corporate and municipal bonds, debentures, notes and commercial paper) and preferred stock rated by
MOODY'S have, prior to assignment of any rating, agreed to pay to MOODY'S for appraisal and rating services rendered by It fees ranging from $1,500 to $2,400,000, Moody's Corporation
(MCO) and ItS wholly-owned credit rating agency subsidiary, Moody's Investors Service (MIS), also maintain policies and procedures to address the Independence of MIS's ratings and rating
processes, Information regarding certain affiliations that may eXist between directors of MCO and rated entities, and between entities who hold ratings from MIS and have also publicly
reported to the SEC an ownership Interest In MCO of more than 5%, IS posted annually on Moody's website at www,moodys,com under the heading "Shareholder Relations - Corporate
Governance - Director and Shareholder Affiliation Policy"
Moody's Investors Service Pty Limited does not hold an Australian financial services licence under the Corporations Act. This credit rating opinion has been prepared without taking Into
account any of your obJectives, financial situation or needs, You should, before acting on the OpiniOn, consider the appropriateness of the opinion having regard to your own o~ectlves,
financial situation and needs
Special Comment
May 2006
Contact
Phone
New York
Jian Hu
Richard Cantor
1.212.553.1653
The Performance of Structured Finance Ratings:
Full-Year 2005 Report
Highlights
This Special Comment updates Moody's structured finance rating performance metrics as of year-end 2005, both with
respect to rating accuracy and rating stability.
As indicated in Figure 1, the performance of the most recent annual rating cohort ending December 2005
improved sharply by almost all measures, relative to six and twelve months prior and the historical average. The highlights of this report show that:
• The most recent cohort's one-year accuracy ratio (AR) was 88.7%, up sharply from its level six months
prior and much higher than its long-run historical average.
• The five-year horizon AR for the most recent cohort was 64.1 %, below both its level six months earlier and
its historical average.
• The one-year investment-grade loss rate was 0.03 %, down substantially from its levels six months prior and
significantly below its historical average.
• The 36-month average rating of all securities that become impaired over a 12-month period was Ba3, two
notches lower than the average rating six months prior and its long-term historical average.
• The rating action rate was 8.1 %, about the same as the long-term historical average and lower than both its levels
six and twelve months prior, as increases in upgrade activity continued to offset declines in downgrade activity.
• The large rating action rate (rating changes of three notches or more) was 3.3%, considerably lower than
the rates observed in the recent past and its historical average.
• Improvements in rating performance were broadly based, resulting in rating accuracy in all sectors during
the last year that exceeded their respective long-term historical averages.
These performance metrics should, however, be interpreted with caution.
• Some statistics are based on small samples, as the number of impairments in any given year and particularly
within any given sector is often small. Individual performance metrics may also be driven by highly correlated collateral performance underlying multiple securitizations, particularly within certain sectors.
• Variation in ratings performance over time may reflect either changes in the quality of the ratings process or
changes in the environment that make defaults more or less difficult to predict or make collateral performance more volatile, thus resulting in either higher rating accuracy and lower rating stability at one time
than another, or vice versa.
Moody's Investors Service
Global Credit Research
Figure 1 - Summary of All Structured Finance Rating Performance as of Year-end 2005*
Cohort Ending Date
Number of
Newly
Impaired
Tranches
over Prior 12
Months
December 2005
June 2005
December 2004
June 2004
Average (1993-Most Recent)
93
143
213
239
84
......................... Accuracy Measures .........................
1·Year
Accuracy
Ratio
5·Year
Accuracy
Ratio
1·Year
Investment·
Grade loss
Rate
36· Month·
Average
Rating
Before
Impairment
83.7%
76.3%
73.0%
71.5%
75.5%
64.1%
65.5%
67.3%
71.0%
70.7%
0.03%
0.10%
0.24%
0.35%
0.27%
Ba3
Bal
Baa3
Bal
Bal
... Stability Measures ...
1·YearRating
Action Rate
1· Year large
Rating
Action Rate
8.1%
9.9%
9.4%
9.4%
8.2%
3.3%
4.5%
4.7%
4.9%
4.1%
Figure 2 - Summary of Structured Finance Rating Performance by Sector as of Year-End 2005*
......................... Accuracy Measures .........................
Number of
Newly
Impaired
Tranches
over Prior 12
Months
Cohorts Ending December 2005
USABS
USCMBS
US RMBS
Global COOs
Int'! SF ex. COOs
Historical Averages Since 1993
USABS
USCMBS
US RMBS
Global COOs
Int'I SF ex. CDOs
... Stability Measures ...
36· Month·
Average
Rating
Before
Impairment
1.YearRating
Action Rate
1· Year large
Rating
Action Rate
1·Year
Accuracy
Ratio
5·Year
Accuracy
Ratio
1·Year
Investment·
Grade loss
Rate
40
32
11
15
0
96.0%
87.5%
87.0%
72.5%
na
62.7%
82.5%
90.6%
56.7%
59.1%
0.01%
0.00%
0.02%
0.10%
0.00%
Ba2
B2
Ba3
Bal
na
4.8%
19.9%
7.7%
7.2%
7.4%
2.3%
5.6%
4.1%
3.1%
2.8%
42
7
12
23
0*
76.7%
86.5%
82.0%
71.0%
61.9%
72.1%
86.3%
75.1%
61.6%
83.4%
0.37%
0.04%
0.09%
0.82%
0.01%
Baa3
B2
Ba2
Ba2
Baa2
7.1%
11.7%
7.1%
12.9%
5.6%
4.4%
3.7%
3.4%
7.1%
1.9%
¥ A glossary appears at the end of this report. The historical average of the number of newly impaked tronches over prior 1Z months is the total number of
impakments divided by the number of years in the sample. and has been rounded to integers, For the international structured finance excluding COOs (Int'l SF
ex. COOs) sector, there arejust five impairments in the entire sample, resulting in an average of O. 4 new impairments per year,
2
Moody's Special Comment
Table of Contents
Introduction .................................................................................................................................... 4
Accuracy Ratios .............................................................................................................................. 5
Investment-Grade Loss Rates ......................................................................................................... 6
Average Rating Before Impairment ................................................................................................. 7
Rating Action Rates and Large Rating Action Rates ......................................................................... 8
Glossary ....................................................................................................................................... 13
Related Reaseach ......................................................................................................................... 15
Moody's Special Comment
3
Introduction
In a Special Comment published in April 2003, Moody's developed a set of metrics to measure the performance of cor-
porate ratings with respect to the dual objectives of rating accuracy and rating stability. 1 Moody's corporate rating performance report is now updated on a quarterly basis. 2 Moody's first introduced and examined its structured finance
ratings performance metrics in "Default & Loss Rates of Structured Finance Securities: 1993-2003," a Special Comment, dated September 2004. In September 2005, Moody's reported comprehensive rating performance metrics for the
first time in a single document. This Special Comment is the second such comprehensive rating performance report for
structured finance globally.
As was the case in the September-2005 report, the basic unit of observation in this report is a monthly cohort of
ratings, i.e., all outstanding ratings at the beginning of a month, and the metrics compare the performance of different
rating cohorts over time. Furthermore, we adjust the performance metrics by loss severity because Moody's structured
ratings are intended to rank order expected loss rates, not simply expected default rates. In other words, in computing
rating performance metrics for structured finance, we intend to weigh those tranches that have become materially
impaired but with lower loss severity less than those with higher loss severity.
The most important measure of rating accuracy is the accuracy ratio, which for structured finance measures the
relationship between security ratings and their realized loss rates. 3 This metric measures the quality of Moody's ratings
as indicators of relative expected credit loss risk.
As discussed in the April 2003 Special Comment, although relative rating accuracy is our primary objective, Moody's
recognizes that many investors are also concerned with the cardinal accuracy of the rating system. In particular, they
expect that investment-grade credits should rarely suffer credit losses and impaired securities should normally carry
low ratings well in advance of impairment. For this purpose, we will regularly track investment-grade loss rates and the
average rating of securities during the 36 months prior to impairment. Both of these measures should be low if the rating system is accurate in a cardinal sense.
We employ two measures of rating stability (or rating volatility) - the rating action rate (the frequency of rating
changes) and the large rating action rate (the frequency of rating changes of three notches or more) over a twelvemonth period. To examine how watchlist actions are used, we also report the percentages of downgrades and upgrades
preceded by watchlist (review) actions.
We calculate accuracy statistics based on the data from Moody's latest structured finance default and loss study
(1993-2005).4 For the measurement of stability and volatility, we incorporate the most recent rating action data from
Moody's latest structured finance rating transition study. 5 In particular, the data sample of this report includes all public, 144A, and private tranches with a Moody's long-term global debt rating in the global asset-backed securities (ABS),
commercial and residential mortgage-backed securities (CMBS and RMBS), and collateralized debt obligations
(CDO) sectors. Prospective ratings, credit evaluations, and national scale ratings are not included. Pari passu tranches
are collapsed into a single tranche. In addition, the following tranches are excluded: (1) tranches guaranteed by financial guarantors or government-sponsored-enterprises (GSEs); (2) interest-only (10) or residual tranches; (3) repackaged securities, structured notes, structured investment vehicles, structured covered bonds, and other credit derivative
securities; and (4) deals that have all their tranche ratings linked to a single corporate or sovereign rating.
1.
2.
3.
4.
5.
4
See "Measuring the Performance of Corporate Bond Ratings." Moody's Special Comment. April 2003.
For the latest performance report. see "The Performance of Moody's Corporate Bond Ratings: March 2006 Quarterly Update." Moody's Special Comment. April 2006.
The required adjustments to convert the standard default-based AR measure to a loss rate-based measure are discussed in "Default & Loss Rates of Structured
Finance Securities: 1993-2003." Moody's Special Comment, September 2004. The accuracy ratio is also described in the Glossary at the end of this report.
See "Default & Loss Rates of Structured Finance Securities: 1993-2005," Moody's Special Comment, April 2006.
See "Structured Finance Rating Transitions: 1983-2005," Moody's Special Comment, February 2006.
Moody's Special Comment
Accuracy Ratios
Figure 3 depicts the one- and five-year accuracy ratios over time with the following notable observations:
• The one-year accuracy ratio in the all structured finance category trended up significantly in 2005, led primarily by the dramatic improvement in the US ABS and US RMBS sectors. In the US ABS sector, the oneyear accuracy ratio jumped to 96.0% in the latest annual cohort ending December 2005, from 80.3% six
months prior and 75.2% twelve months prior. Similarly in the US RMBS sector, the one-year accuracy
ratio jumped to 87.0% from 76.1 % and 62.5%.
• The accuracy ratio in the US CMBS sector remained high and was above its historical average of 86.5% in
the most recent annual cohort.
• The one-year accuracy ratio of Global CDOs also improved substantially to 72.5% in the latest annual
cohort, from 62.8% and 67.2% six and twelve months earlier.
Figure 3 - One-Year (solid line) and Five-Year (dotted line) Accuracy Ratios
All Structured Finance
100%
USABS
?~~
90%
80%
70%
60%
50%
100%
90%
80%
70%
60%
50%
40%
40%
30%
30%
20%
20%
10%
10%
0%
0%
-------
~
1
~ ~ j
~
-------
USCMBS
~
90%
iI
70%
I
60%
~ ~ j
~
~\(lIy
90%
80%
70%
1
50%
'.V'"I
30%
I
40%
30%
100%
60%
J
50%
1
USRMBS
100%
80%
~
40%
20%
20%
10%
10%
0%
0%
-------
~
1
~ ~ j ~
-------
Global CDOs
~
1
~ ~ j
~
Inl'l SF ex. CDOs
100%
100%
90%
90%
80%
80%
70%
...
60%
~
.
. .
70%
50%
50%
40%
40%
30%
30%
20%
20%
10%
\
J
..
60%
10%
0%
0%
8
;J;
ill
ill
:;;
ill
ill
:5
5
i'i
8
11
:'l
-------------
8
;J;
ill
ill
:;;
ill
g:
:5
5
i'i
8
11
:'l
--------- ----
Note: At the beginning of each month, securities are grouped together by their alpha-numeric ratings to form rating cohorts. The latest one- and fiveyear cohorts are formed on January 1, 2005 and January 1, 2001.
The gaps in the US CMBS and International Structured Finance accuracy ratio series resulted from zero impairment in those periods. In addition, the
accuracy ratios tend not to be meaningful when the number of impaired securities is very small (in single digits). For example, the five-year accuracy
ratio for international structured finance is based on a single impaired security that was rated Ba2 in 1996 and impaired in 2002. This structured finance
sector had just five impaired tranches, four of which were rated after January 1, 2000. The impairment experiences of these four tranches are reflected
in the one-year accuracy ratios. In addition, only one CMBS security was impaired in 1994, and was rated Baa2 before it became impaired.
Moody's Special Comment
5
Investment-Grade Loss Rates
Figure 4 shows that the loss rates of investment-grade securities have declined significantly across all sectors in the latest annual cohort ending December 2005,6 whereas the loss rate in the latest five-year cohort ending December 2005
was still at its historical high in the all structured finance category as well as in the US ABS and Global CDOs sectors?
In addition,
• In the all structured finance category, the investment-grade loss rate dropped to 0.03% in the latest annual
cohort from 0.10% and 0.24% six and twelve months prior.
• The US ABS and Global CDO sectors experienced significant declines in the loss rates, with the one-year
US ABS investment-grade loss rate decreasing to 0.01 % in the latest annual cohort from 0.38% one year
prior, and to 0.10% from 0.58% over the same one-year period for Global CDOs.
• The investment-grade loss rates remained very low in the US CMBS and RMBS sectors. In particular, the
US CMBS investment-grade loss rate has been at 0% since August 2003 and during most of the sample
period, and during 2000-2002 for US RMBS.
Figure 4 - One-Year (solid line) and Five-Year (dotted line) Investment-Grade loss Rates
All Structured Finance
USABS
2.5%
3.5%
3.0%
2.0%
2.5%
1.5%
2.0%
1.5%
1.0%
~~
I
0.5%
I
I
I
0.0%
...
C'-;=o
~
1.0%
0.5%
0.0%
g:
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
US CMBS
2.5%
2.0%
~
2.0%
1.5%
~
1.5%
"~ '~"
~
8;
~
ell
~
~
~
5
"~ "~
1\Z~~n
I
~
0.0%
~
.
0.5% -(
ill
~
~
~
g
>'l
,
1.0%
0.5%
g:
~
US RMBS
2.5% -
'"i ~
~
'~" "~
5
~
"~ "~
0.0%
g
~
,
I
g:
>'l
~
ill
"~ '~"
~
~
Global COOs
'~" "~
ell
8;
~
~
~
~
~
Int'l SF ex. COOs
10.0%
2.5%
9.0%
2.0%
8.0%
7.0%
1.5%
6.0%
5.0%
1.0%
4.0%
3.0%
0.5%
2.0%
1.0%
0.0%
0.0%
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
Note: At the beginning of each month, securities carrying an investment-grade rating are grouped together to form a rating cohort. The latest one-year
rating cohort is formed on January 1, 2005 and the latest five-year rating cohort is formed on January 1, 2001.
6.
7.
6
These loss rates are based on number of securities, not dollar volume of issuance. Loss rates weighted by dollar volume are substantially lower.
Differences in rating distribution within the investment-grade category may contribute to the differences of investment-grade loss rates across sectors. For more details
about the impairment and loss rates of structured finance securities by rating and sector, please see "Default & Loss Rates of Structured Finance Securities: 19932005," Moody's Special Comment, April 2006.
Moody's Special Comment
Average Rating Before Impairment
Figure 5 presents the 36-month-average rating before impairment over time. 8 To provide the context of this metric,
the figure also contains the number of newly impaired securities used to calculate the average ratings over time. The
following observations are noteworthy:
• The average rating before impairment in the all structured finance category dropped sharply by three notches
over the past year, to roughly Ba3 in the most recent cohort from Baa3 twelve months prior. This was led primarily by the declines in the average ratings before impairment in the US ABS and US RMBS sectors.
• The average rating before impairment in US CMBS continued to be the lowest at B2 among all sectors in
the latest annual cohort, only slight higher than B3 one year earlier.
• The average rating before impairment in the Global CDO sector was at Bal in the most recent annual
cohort, about the same as those six and twelve months prior, and one notch higher than the historical average ofBa2.
• The average ratings before impairment in the US ABS and RMBS sectors have been volatile historically,
especially during periods in which the number of newly impaired securities was small. WIthin cohorts ending in 2005, the average rating before impairment in the US ABS sector has declined significantly and
approached Ba2 in the most recent cohort. The Ba2 rating was three notches lower compared to one year
earlier. In addition, the decline in the average rating before impairment for US RMBS was equally impressive, falling to Ba3 in the latest annual cohort from Baa2 twelve months prior.
Figure 5 - 36-Month-Average Ratings before Impairment (solid line) and
Number of Newly Impaired Securities (dotted line)
USABS
All Structured Finance
Baal
500
Baa2
450
400
Baa3
350
B,l
250
Baal
Baa2
200
Baa3
B,l
150
300
B,2
250
B,2
B,3
200
B,3
100
150
B1
B1
50
B2
50
B2
B3
£i
£i
£i
£i
£i
£i
£i
iii
£i
q
£i
q
iii
£i
£i
~
B3
q
USCMBS
1
~ ~ j ~
USRMBS
Baal
100
Baa2
Baa3
80
Baal
100
Baa2
80
Baa3
B,l
60
B,2
B,3
B3
...., .......
Caal
-------
1
40
B,3
20
B1
20
B2
'
~ ~
60
40
Caa2
~
B,l
B,2
~ ....
B1
B2
8.
~
-------
£i
~
-'i
~
B3
£i
£i
£i
£i
£i
£i
£i
iii
£i
q
£i
iii
£i
q
£i
'i
£i
q
£i
We have adjusted this metric by the differences in loss severity across impaired tranches. There are a large number of non-matured tranches. the loss severity of
which have to be estimated. and may be subject to revisions once more data become available.
Moody's Special Comment
7
Figure 5 - 36-Month-Average Ratings before Impairment (solid line) and
Number of Newly Impaired Securities (dotted line)
Global CDOs
Baal
Inl'l SF ex. CDOs
14
A1
250
Baal
Baa2
200
Baa3
12
Baa2
10
Baa3
B,l
150
B,2
100
B,3
B,l
B,2
B,3
B1
50
B2
B1
B2
B3
B3
~
~
~
~
". ..... , • • • • • • • • • • • • • • • • • • • • 1
••••••••••••••••••••1
iii
~
~
~
~
~
~
~
~
.....
.~
~
iii
~
......
,.~'
~
'1
~
..
~
Note: At the beginning of each month. all securities that become impaired within the next 12-month period are grouped to form a cohort. The latest
cohort is formed on January 1. 2005.
Rating Action Rates and Large Rating Action Rates
Figure 6 reports annual rating action rates and large rating action rates. Figures 7 and 8 further disaggregate rating
actions into downgrades and upgrades, and Figure 9 demonstrates how frequently downgrades and upgrades have
been preceded by watchlist actions in the same direction. 9 Key observations include:
•
•
•
•
The annual rating action rate in the all structured finance category decreased in the second halfof2005 to 8.1 %
in the most recent annual cohort, from 9.9% six months earlier and 9.4% twelve months earlier, as downgrade
rates continued to decline and upgrade rates started to taper off in 2005. Additionally, the number oflarge rating
actions as a share of all rating actions shrank to 41 % in the latest cohort from 49% a year earlier.
The downgrade rate has declined substantially to 1.9% in the latest one-year cohort ending December
2005, from 3.3% and 5.0% half-a-year and a year prior, respectively. In addition, the large downgrade rate
- downgrades of three notches or more - was halved to 0.9% in the most recent annual cohort, from
1.8% six months earlier.
The upgrade rate was at 6.2%, in the latest annual cohort, slightly lower than the 6.6% level six months
earlier, but much higher than the historical average upgrade rate of 3.9%.
Most downgrades have been anticipated by negative rating reviews, although in recent annual cohorts that
ended in 2005, the frequency of reviews prior to downgrades has declined;lO in the most recent annual
cohort, that frequency was 72.7%, compared to 78.7% a year prior. Additionally, most upgrades were not
reviewed prior to a rating action as only about 41 % of all upgrades in the most recent annual cohort were
anticipated by a positive review, marking a slight decline from 46% a year earlier.
9.
Different rating distributions may contribute to the differences in rating actions rates across sectors. For more details about structured finance rating transitions by rating and sector, please see "Structured Finance Rating Transitions: 1983-2005." Moody's Special Comment. February 2006. Moody's also reviewed its uses of
watchlist actions on structured finance securities in a Special Comment. "Structured Finance Watchlist Resolutions: 1992-2003." June 2004.
10. Almost all 2005 downgrades in the US CMBS sector occurred in the speculative-grade category, most of which were not placed on review before their downgrades.
8
Moody's Special Comment
Figure 6 - Annual Rating Action Rates (dotted line) and
large (three notches or more, solid line) Rating Action Rates
All Structured Finance
USABS
US CMBS
US RMBS
24%
21%
18%
15%
12%
R
.
.
:dS>~
Global COOs
30%
27%
Inl'l SF ex. COOs
24%
21%
24%
18%
21%
18%
15%
12%
15%
12%
9%
9%
6%
6%
3%
O%+-~~--~~--~~~--~--~--~--~--~~--~
3%
•
0%10
Note: Rating actions include upgrades and downgrades, which are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed
each month. The latest annual cohort is formed on January 1, 2005.
Moody's Special Comment
9
Figure 7 - Annual Downgrade Rates (dotted line) and
Annual large (three notches or more, solid line) Downgrade Rates
All Structured Finance
USABS
10%
12%
10%
8%
8%
6%
6%
4%
4%
2%
2%
0%
0%
~
~
~ ~
~
~
~
~
~
~ ~ ~
~
~ ~
~ ~
~
US CMBS
10%
8%
8%
6%
6%
4%
4%
2%
2%
~
; ; ;
~
; ;
~
~
~ ~
~ ~
;
~ ~
US RMBS
10%
0%
~ ~
~
~
;
0%
~ ~
~
~
; ; ;
~
Global COOs
; ;
~
~
~
Inl'l SF ex. COOs
30%
16%
28%
14%
26%
24%
12%
22%
20%
10%
18%
16%
8%
14%
12%
6%
10%
8%
4%
6%
4%
2%
2%
0%
0%
~
~
~
~
~
~
~
~
~
~
~
~
~
iO
~
~
~
b
~
~
~
~
~
~
~
~
~
~
~
Note: Downgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month. The latest annual cohort is
formed on January 1,2005.
10
Moody's Special Comment
Figure 8 - Annual Upgrade Rates (dotted line) and
Annual large (three notches or more, solid line) Upgrade Rates
All Structured Finance
USABS
10%
10%
8%
8%
6%
6%
4%
2%
..
~
;
"
0%
4%
..
~
~
~
~
...
,
2%
'
~
~
~
~
~
~ ~
0%
~
~
~
~
~
US CMBS
~
;
~
~
~
~
~ ~
~
~
~ ~
~
~
~ ~
~
US RMBS
18%
10%
16%
8%
14%
12%
6%
10%
8%
4%
6%
4%
2%
2%
0%
0%
~
~
~
~
~
~
~
~
~
~
~ ~
~
~
~
~
~
Global COOs
~ ~
~
~
~
Inl'l SF ex. COOs
15%
10%
13%
8%
11%
6%
9%
7%
4%
5%
2%
3%
1%
0%
-1%
~
~
~
~
~
~
~
~
~
~
~ ~
~
~
~
~
~
~ ~
~
~
~
Note: Upgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month. The latest annual cohort is formed
on January 1, 2005.
Moody's Special Comment
11
Figure 9 - Percentages of Downgrades (solid line) and
Upgrades (dotted line) Preceded by Watchllst Actions in the Same Direction
All Structured Finance
USABS
100%
100%
90%
90%
80%
80%
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
0%
; ;
~
; ;
~
~
~
;
~
;
0%
~
~
; ; ;
~
; ;
~
~
;
~
;
~
US CMBS
US RMBS
100%
100%
90%
M~
90%
80%
80%
70%
,
70%
60%
II
60%
50%
50%
40%
40%
30%
30%
20%
,!
20%
10%
10%
0%
0%
~
~
~
~
~
~
~
~
~
~
~
~
; ; ;
~
; ;
~
~
~ ~
;
~
~ ~
;
~
Global COOs
Inl'l SF ex. COOs
100%
100%
90%
90%
80%
80%
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
-.-~
0%
10%
,"-"- ,"-"
0%
~
~
~
~
~
~
~
~
~
~
~
~
~
~
; ; ;
~
; ;
~
~
Note: At the beginning of each month, all securities that experienced downgrades (upgrades) within the next 12-month period are grouped together to
form a rating cohort. The latest annual cohort is formed on January 1, 2005.
12
Moody's Special Comment
MATERIAllMPAIRMENT
Material impairment, or for simplicity, impairment, is a concept adopted in Moody's structured finance default and loss
studies. Material impairment includes uncured payment defaults and securities downgraded to Ca or C, where payment defaults include shortfalls of interest and losses of principal on individual tranches of structured finance transactions. A security is called "newly impaired" in a given period if it had no outstanding interest shortfalls, no principal
losses, and was not rated Ca or C in the prior period, but experienced at least one of these three credit events in the
given period for the first time.
lOSS-GIVEN-DEFAUlT (lGD)
Loss-given-default, also known as LGD or loss severity rate, is the total amount of lifetime losses as a share of a
tranche's principal balance on a certain reference date. The losses in each payment period are discounted by a discount
rate, which is typically the stated coupon rate on the tranche. There are three types of principal balances used in the
calculation of LGD: the principal balanced at origination, at the time of impairment, and at a cohort formation date.
Depending on the reference principal balance (and the reference date), the calculated LGD can be significantly different due to principal amortization and discounting.
Final LGD on impaired securities with no outstanding principal balances are typically known. For impaired securities with positive principal balances or impaired securities with incomplete loss data, their LGD need to be estimated. Detailed discussions on the estimation of those final LGD are provided in several Moody's Special Comments
including, in particular, "Measuring Loss Severity Rates of Defaulted Residential Mortgage Backed Securities,"
Moody's Special Comment, April 2004, and "Default & Loss Rates of Structured Finance Securities: 1993-2005,"
Moody's Special Comment, April 2006.
ACCURACY RATIO
An accuracy ratio (AR) based on the loss experience of structured finance securities (or an accuracy ratio adjusted for
loss-given-default, or LGD) is the ratio of the area between a loss-based cumulative accuracy profile (CAP) curve and
the 45-degree line to the maximum possible area above the 45-degree line. The loss-based CAP curve (or a CAP curve
adjusted for LGD) plots, for each rating category, the proportion of the losses of all impaired securities accounted for
by securities with the same or lower rating against the proportion of all securities in the sample population with the
same or lower rating. I I
To calculate accuracy ratios, we use the same data sample that is used in the latest structured finance default and loss
study (1993-2005), and form rating cohorts of different horizons for each calendar month during 1993-2005, i.e. all outstanding securities in the sample population are grouped together by their alpha-numeric ratings at the beginning of the
month. We then do the same (group by ratings in each calendar month) for securities that became impaired within one,
three, or five years from that month. Cumulative shares of securities rank-ordered by rating are calculated for the universe of all securities and the universe of impaired securities, respectively, and based on this, a CAP curve is plotted.
The CAP curve adjusted for LGD is also known as a "power curve" because it shows how effective a rating system
is at differentiating between securities that have sustained high losses from securities that have sustained low losses or
no losses. The metric is defined relative to the distribution of ratings in the population. The accuracy ratio is an effective way to summarize the CAP curve into a single number. The accuracy ratio is zero if the CAP curve collapses to the
45-degree line, suggesting that all impaired securities are randomly distributed throughout the population without
regard to ratings.
Because we use only data observations that span the time horizon under consideration, the latest cohort for a oneyear horizon is formed onJanuary 1, 2004, while for a five-year horizon the latest cohort is formed onJanuary 1, 2000.
INVESTMENT -GRADE lOSS RATE
The one-year investment-grade (Aaa through Baa3) loss rate is calculated as follows. First, for each impaired security,
we calculate a loss rate, given impairment, as a share of its tranche balance at the cohort date. Then, we sum up these
loss rates to get a total loss value for all securities that become impaired within a 12-month period after the cohort formation date. We divide this total loss value by the total number of securities in the rating cohort to obtain the invest11. For an illustration of the CAP curve adjusted for LGD. see "Defauft & Loss Rates of Structured Finance Securities: 1993-2003." Moody's Special Comment. September 2004.
Moody's Special Comment
13
ment-grade loss rate. Note that this loss rate is not weighted by the total dollar volume of outstanding securities in the
cohort. The five-year investment-grade loss rate is calculated similarly.
AVERAGE RATING BEFORE IMPAIRMENT
The rating of an impaired security is measured every month for 36 months prior to impairment as well as immediately
prior to impairment. These 37 rating measurements are averaged together to create one representative number for
each impaired security that becomes impaired within 12 months after the month of cohort formation. This average is
weighted by each impaired tranche's loss severity rate as a share of original balance so that the rating on an impaired
tranche with higher loss severity rate is weighted more than the rating on a tranche with lower loss severity rate. 12
These representative numbers are then averaged together to create the reported average rating each month.
RATING ACTION RATE
The rating action rate is defined as the number of securities that experienced a rating change within a year after cohort
formation divided by the total number of securities outstanding at the cohort formation date (the beginning of each
month). Rating changes are measured on the alpha-numeric (or modified) rating scale. Downgrade rates and upgrade
rates are measured similarly based on downgrade rating actions and upgrade rating actions, respectively.
LARGE RATING ACTION RATE
A security that had its rating changed by three notches or more within a year after cohort formation is said to have
experienced a large rating action. The large rating action rate is the number of such securities divided by the total
number of securities outstanding at the cohort formation date. Large downgrade rates and large upgrade rates are
measured similarly based on large downgrade rating actions and large upgrade rating actions, respectively.
PERCENTAGE OF DOWNGRADES (UPGRADES) PRECEDED BY WATCH LIST ACTIONS
This metric is defined as the number of securities that were placed on the watchlist in the same direction before they
were downgraded (upgraded), divided by the total number of securities that were downgraded (upgraded) within 12
months after the cohort formation date.
ABS
This refers to Asset-Backed-Securities. This structured finance sector includes securities backed by home equity loans
(HEL) in addition to both traditional (autos, credit cards, leases, manufactured housing, student loans, etc.) and nontraditional (mutual fund fees, tax liens, tobacco settlement, whole business securitizations (ABS), etc) asset classes.
HELS
HELs include securities collateralized by subprime (B&C) mortgage loans, home improvement loans, high loan-tovalue (high LTV) loans, home equity lines of credit (HELOs), closed-end second-lien loans, and net interest margin
(NIM) securitizations. It does not include "Alt-A" mortgages, which are part of the RMBS sector. HEL is part of the
ABS sector.
CDOS
This refers to collateralized debt obligations. Credit derivative securities such as repackaged securities and structured
notes are not considered to be part of this sector.
CMBS
This refers to commercial mortgage-backed securities.
12. In a simplified example of two impairments: one with a loss severity rate of LGD(1) and an average rating number of RN(1). and the other with a LGD(2) and RN(2).
the average rating is calculated as (LGD(1 tRN(1) +L GD(2tRN(2))/(L GD(1) +LGD(2)). This is the first time we introduce LGD rates as weights in computing an average rating before impairment. Ideally, LGD rates should be calculated for each monthly principal balance in the 36 months prior to the impairment. and ratings should
then be weighted by these monthly LGD rates. Practically, however, it does not make any material difference in the average rating number, whether we use LGD rates
as a share of impairment-date balance or original balance or the monthly LGD rates. Since the monthly LGD rates are very time consuming to compute due to amortization. we use the LGD rate as a share of original balance as the weight variable.
14
Moody's Special Comment
RMBS
This refers to residential mortgage-backed securities. The large majority of these securities are backed by first-lien
prime mortgages, although some are backed by Alt-A mortgages. In some older vintage RMBS transactions, subprime
mortgages are also included in the collateral. HEL is not part of this sector.
U.S. STRUCTURED FINANCE SECURITIES
This refers to structured finance securities denominated in U.S. dollars and issued in the U.S. market.
INTERNATIONAL STRUCTURED FINANCE EXCLUDING CDOS
This refers to non-CDO structured finance securities that are not denominated in U.S. dollars or not issued in the
U.S. market. The majority of the securities in this sector are in Europe; the rest comes from the Asia Pacific region,
Canada, and Latin America.
Related Research
Special Comments:
The Performance of Structured Finance Ratings: Mid-Year 2005 Report, September 2005 (94463)
P~f,!p-:!t~..1Q~?R,!!.~?..Qf.~!!J!~~X.~g.X!!!~!!.~~.~~.~~x.hj~~:J.2~~.:2:Q.Q.~,.A2IH.IQQ.§..C2Z.n.12
R~.f'!~J.!..~.1.Q~~..R~!~~..Q.f~!T~f.~r~.Q..f~!l~!l~~..~.~~.\;!r!!!.~?:J.2.21.::~.QQ:h.lp-:k~.QQ.H~~.§~.n
P~f,!p-:!t~..1Q~?R,!!.~?.. Q.f.Y:~:..~.PQ~:)~2}::~.Q.Q}LM~r~h.IQQ?..(n.§~n
R~.f'!~J.~..~.1.Q~~ ..R~!~~..Qf.~~.c;:.~~.Q.f~!l'!!l~~..~.~~~r!!!.~.?:lQ.Q±Jnx.?qi~J.f.V'p'Q~.~~-,.l~!!~'!IT.lQQ.?'..(~Q.~:m
P~f,!p-:!t~..1Q~?R,!!.~?.. Qf.~.~!J!~~X.~g.X!!!.'!!!.~~ ..~~.~~X.hj~~:J.22J:.~Q.Q},5~p.~~.~g.~rlQ.Q±J~.~g.2l)
Measuring Loss Severity Rates of Defaulted Residential Mortgage Backed Securities, April 2004 (86769)
Payment Defaults and Material Impairments of U.S . Structured Finance Securities: 1993-2002, December 2003 (80247)
Structured Finance Rating Transitions: 1983-2005, February 2006 (96533)
Structured Finance Rating Transitions: 1983-2004, February 2005 (91392)
.~~~~!~.Q.X!!l~!!f.~.R'!!:~!lg.Tr~!!~!!!.Q!!~:)~~.~::~.Q.Ql2.f~g.~~.ry.~.Q.Q±.£~J.n~)
~.~f~I~QXj!!.'!!l~~..R~!j!!K=n·.'!!l?~!:~Q!l?:.J.2.~l.:.~QQ.~,..~.Q!nl?'!r~~Q!!.?~.!h..~.Qr2Qr'!!.~.R~!:!!lg?..~!lg..AfX.Q~~..~~.~!.Qr~,
1~!lp-:,!ry..~QQL(ZZ2:2.!2
.~~~~!~.Q.X!!!~!!f.~.}Y'!.!fh!!~tR~~Q!p-:!:!Q!!~:J.2~~.:2:QQ1.l.1~~.IQQ.1..c~Z.~Q?2
Q~!.M.~.Q.M.QQgi?R~.f'!~k.R~.?.~'!r~h:.f~QP-:,!ry.IQQ.§..v'p.Q.'!.~.~-,..M~!.~h.IQQ.g..(~@±!2
Th.~..~~x.fQr~~!!f.~.Qf.M.Q.QQi~ ..~Qr2Qr'!.!.~..J?Q!lg..R'!.~.iDg~:..M.'!r~h.~.Q.Q§'.Q.~.'!.It:~X!Y..Y2Q.'!.~.~L~.Px.H.IQQ.§..(2.Z~l~2
Measuring the Performance of Corporate Bond Ratings, April 2003 (77916)
To access any of these reports, click on the entry above. Note that these references are current as of the date ofpublication ofthis repo1't
and that more recent rep017S may be available. All research may not be available to all clients.
Moody's Special Comment
15
To order reprints of this report (100 copies minimum), please call 1.212.553.1658.
Report Number: 97346
Author
Production Specialist
Jian Hu
Yung Louie
© Copyright 2006, Moody's Investors Service, Inc, and/or Its licensors and affiliates including Moody's Assurance Company, Inc, (together, "MOODY'S"), All rights reserved, ALL
INFORMATION CONTAINED HEREIN IS PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR OTHERWISE REPRODUCED, REPACKAGED,
FURTHER TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN WHOLE OR IN PART. IN
ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER. BY ANY PERSON WITHOUT MOODY'S PRIOR WRITTEN CONSENT. All information contained herein IS obtained by
MOODY'S from sources believed by It to be accurate and reliable, Because of the possibility of human or mechanical error as well as other factors, however, such information IS provided "as
IS" without warranty of any kind and MOODY'S, In particular, makes no representation or warranty, express or Implied, as to the accuracy, timeliness, completeness, merchantability or fitness
for any particular purpose of any such Information, Under no circumstances shall MOODY'S have any liability to any person or entity for (a) any loss or damage In whole or In part caused by,
resulting from, or relating to, any error (negligent or otherwise) or other circumstance or contingency within or outside the control of MOODY'S or any of Its directors, officers, employees or
agents In connection with the procurement, collection, compilation, analysIs, Interpretation, communication, publication or delivery of any such information, or (b) any direct, indirect,
special, consequential, compensatory or incidental damages whatsoever (Including without limitation, lost profits), even If MOODY'S IS advised In advance of the possibility of such
damages, resulting from the use of or inability to use, any such information, The credit ratings and financial reporting analysIs observations, If any, constituting part of the information
contained herein are, and must be construed solely as, statements of opinion and not statements of fact or recommendations to purchase, sell or hold any SeCUrities, NO WARRANTY,
EXPRESS OR IMPLIED, AS TO THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR OTHER
OPINION OR INFORMATION IS GIVEN OR MADE BY MOODY'S IN ANY FORM OR MANNER WHATSOEVER. Each rating or other opinion must be weighed solely as one factor In any
Investment decIsion made by or on behalf of any user of the information contained herein, and each such user must accordingly make ItS own study and evaluation of each security and of
each Issuer and guarantor of, and each provider of credit support for, each security that It may consider purchasing, holding or seiling
MOODY'S hereby discloses that most Issuers of debt seCUrities (including corporate and municipal bonds, debentures, notes and commercial paper) and preferred stock rated by
MOODY'S have, prior to assignment of any rating, agreed to pay to MOODY'S for appraisal and rating services rendered by It fees ranging from $1,500 to $2,400,000, Moody's Corporation
(MCO) and ItS wholly-owned credit rating agency subsidiary, Moody's Investors Service (MIS), also maintain policies and procedures to address the Independence of MIS's ratings and rating
processes, Information regarding certain affiliations that may eXist between directors of MCO and rated entities, and between entities who hold ratings from MIS and have also publicly
reported to the SEC an ownership Interest In MCO of more than 5%, IS posted annually on Moody's website at www,moodys,com under the heading "Shareholder Relations - Corporate
Governance - Director and Shareholder Affiliation Policy"
Moody's Investors Service Pty Limited does not hold an Australian financial services licence under the Corporations Act. This credit rating opinion has been prepared without taking Into
account any of your obJectives, financial situation or needs, You should, before acting on the OpiniOn, consider the appropriateness of the opinion having regard to your own o~ectlves,
financial situation and needs
Special Comment
September 2006
Contact
Phone
New York
Jian Hu
Julia Tung
Richard Cantor
1.212.553.1653
The Performance of Structured Finance Ratings:
Mid-Year 2006 Report
Highlights
This Special Comment updates Moody's structured finance rating performance metrics as of June 2006, both with
respect to rating accuracy and rating stability.
As indicated in Figure 1, the one-year rating accuracy and stability of the most recent rating cohort (ending June
2006) was greater compared to the performance of cohorts formed six and twelve months prior and compared to the
long-term average performance of all prior cohorts. The highlights of this report are:
• Overall, 72 structured finance securities became impaired in the first half of 2006: 31 in US ABS, 29 in US
CMBS, 3 in US RMBS, 8 in global CDOs, and one in the international structured finance sector excluding
CDOs.
• The one-year accuracy ratio for the most recent cohort was 95.2%, which is 5.6 percentage points higher than its
six months-prior level and a dramatic 18.6 percentage points higher than the long-run historical average.
• The five-year accuracy ratio for the most recent five-year cohort, formed in July 2001, was 64.3 % -- below
its level of 66.3 % six months earlier and its historical average of 70.9% -- as this statistic has not yet captured the more recent improvements in performance.
• Almost all newly impaired securities over the past year were rated speculative-grade well in advance of
impairment. As a result, the one-year investment-grade loss rate was 0.01 %, lower than its level of 0.02%
six months ago and a steep decline from its long-term historical average of 0.22%.
• The average rating during the three years prior to impairment of all securities that became impaired over
the past year dropped to an all-time low of B2, two notches below the average rating prior to impairment
six months prior and four notches below the historical average.
• The one-year rating action rate declined to 6.9% in the most recent cohort from its six months-prior level
of 8.0% and its historical average of 8.1 %, as the frequencies of both upgrades and downgrades declined.
• The one-year large rating action rate (rating changes of three notches or more) was 2.8%, lower than the
six months-prior level of 3.2 %, and significantly lower than the historical average of 3.9%.
• Improvements in the one-year accuracy ratios, investment-grade loss rates and average ratings before
impairment were observed across all sectors of structured finance, with global CDOs and US RMBS exhibiting the most dramatic improvements relative to recent historical performance. 1
1.
These performance metrics should be interpreted with caution. Some statistics are based on small samples. as the number of impairments in any given year and any
given sector is often small. Individual performance metrics may also be driven by highly correlated collateral performance under/ying multiple securitizations. particular/y within certain sectors. Moreover, variations in rating performance over time may reflect either changes in the quality of the rating process or changes in the environment that make defaults more or less difficult to predict or make collateral performance more volatile. thus resulting in either higher rating accuracy and lower rating
stability at one time than another, or vice versa.
&,Wi;1 Moody's Investors Service
'§i~? W Global Credit Research
Figure 1 - Summary of All Structured Finance Rating Performance as of June 2006*
......................... Accuracy Measures .........................
Cohort Ending Date
Jul1e 2006
December 2005
Jul1e 2005
December 2004
Average (1993-Most Recent)
Number of
New
Impairments
over Prior 12
Months
106
90
142
208
35
'·Year
Accuracy
Ratio
5·Year
Accuracy
Ratio
95.2%
39.6%
77.6%
75.4%
76.6%
64.3%
66.3%
68.0%
69.9%
70.9%
1·Year
Investment·
Grade loss
Rate
0.01%
0.02%
0.12%
0.21%
0.22%
36· Month·
Average
Rating
Before
Impairment
B2
Ba3
Bal
Baa3
Bal
... Stability Measures ...
1.YearRating
Action Rate
6.9%
8.0%
9.9%
9.4%
8.1%
, . Year large
Rating
Action Rate
2.8%
3.2%
4.5%
4.6%
3.9%
Figure 2 - Summary of Structured Finance Rating Performance by Sector as of June 2006*
......................... Accuracy Measures .........................
Number of
New
Impairments
over Prior 12
Months
Cohorts Ending June 2006
USABS
USCMBS
US RMBS
Global COOs
111t'1 SF ex. COOs
Historical Averages Since 1993
USABS
USCMBS
US RMBS
Global COOs
111t'1 SF ex. COOs
47
40
6
12
42
8
11
23
... Stability Measures ...
1·Year
Accuracy
Ratio
5·Year
Accuracy
Ratio
1·Year
Investment·
Grade loss
Rate
36· Month·
Average
Rating
Before
Impairment
96.2%
89.7%
95.7%
93.5%
99.6%
63.3%
83.6%
92.3%
57.1%
56.6%
0.01%
0.00%
0.00%
0.02%
0.00%
B1
B3
Ba2
B3
B1
4.0%
21.4%
5.1%
5.9%
5.5%
2.1%
7.2%
2.5%
2.7%
1.5%
77.7%
72.7%
85.9%
73.9%
613%
74.6%
0.29%
0.03%
0.08%
0.64%
0.01%
Baa3
B2
Ba2
Ba2
Baa2
6.8%
12.6%
7.0%
12.0%
5.7%
4.1%
4.0%
3A%
6.5%
1.8%
87A%
82.6%
72.6%
59.1%
1·YearRating
Action Rate
1· Year large
Rating
Action Rate
, A glossary appears at the end of this report The number of impairments for Nstorical cohorts is suty"ect to revision during each update as payment defaults can
be cured and past remittance or trustee reports may be revised. The historical average ofthe number of new impairments over the prior 12 months is calculated
as the total number of newly impaired tranches divided by the number of years in the sample period, and has been rounded to the nearest integer.
2
Moody's Special Comment
Table of Contents
Page
Introduction .................................................................................................................................... 4
Accuracy Ratios .............................................................................................................................. 5
Investment-Grade Loss Rates ......................................................................................................... 6
Average Rating Before Impairment ................................................................................................. 7
Rating Action Rates and Large Rating Action Rates ......................................................................... 8
Glossary ....................................................................................................................................... 13
Related Research ......................................................................................................................... 15
Moody's Special Comment
3
Introduction
In a Special Comment published in April 2003, Moody's developed a set of metrics to measure the performance of cor-
porate ratings with respect to the dual objectives of ratin~ accuracy and rating stability. 2 Moody's corporate rating
Moody's first introduced and examined its structured
performance report is now updated on a quarterly basis.
finance rating performance metrics in a September 2004 Special Comment "Default & Loss Rates of Structured Finance
Securities: 1993-2003," and published these performance metrics in a stand-alone document for the first time in September 2005. The structured finance rating performance report is now updated on a semi-annual basis.
For both the corporate and structured finance rating performance reports, the basic unit of observation is a monthly
cohort of ratings, i.e. all outstanding ratings at the beginning of a month are recorded and their performance tracked over
different time horizons. In computing rating performance metrics for structured finance, Moody's incorporates both the
default and loss severity experience of all structured finance tranches because Moody's structured finance ratings rank
order expected loss rates. In other words, Moody's structured rating performance metrics weigh those tranches that have
become materially impaired but with lower loss severity less than those with higher loss severity.
The most important measure of rating accuracy is the accuracy ratio, which for structured finance measures the
relationship between tranche ratings and their realized loss rates. 4 This metric measures the quality of Moody's ratings as indicators of relative expected credit loss risk.
As discussed in the April 2003 Special Comment, although relative rating accuracy is our primary objective,
Moody's recognizes that many investors are also concerned with the cardinal accuracy of the rating system. In particular, they expect that investment-grade credits should rarely suffer credit losses and impaired securities should normally carry low ratings well in advance of impairment. For this purpose, we will regularly track investment-grade loss
rates and the average rating of securities during the 36 months prior to impairment. Both of these measures should be
low if the rating system is accurate in a cardinal sense.
We employ two measures of rating stability (or rating volatility) - the rating action rate (the frequency of rating
changes) and the large rating action rate (the frequency of rating changes of three notches or more) over a twelvemonth period. To examine how watchlist actions are used, we also report the percentages of downgrades and upgrades
preceded by watchlist (review) actions in the same direction.
Accuracy statistics are calculated based on the data from Moody's latest annual structured finance default and loss
study (1993-2005)5 supplemented with default and loss data from the first half of 2006. For the measurement of stability and volatility, we incorporate the most recent rating action data from Moody's latest structured finance rating
transition study. 6 In particular, the data sample of this report includes all public, 144A, and private tranches with a
Moody's long-term global debt rating in the global asset-backed securities (ABS), commercial and residential mortgage-backed securities (CMBS and RMBS) and collateralized debt obligations (CDO) sectors. Provisional ratings,
credit evaluations, and national scale ratings are not included. Pari passu tranches are collapsed into a single tranche.
In addition, the following tranches are excluded: (1) tranches guaranteed by financial guarantors or government-sponsored enterprises (GSEs); (2) interest-only (10) or residual tranches; (3) repackaged securities, structured notes, structured investment vehicles, structured covered bonds, and other credit derivative securities; and (4) deals that have all
their tranche ratings linked to a single corporate or sovereign rating.
2.
3.
4.
5.
6.
4
See "Measuring the Performance of Corporate Bond Ratings," Moody's Special Comment, April 2003.
For the latest performance report, see "The Performance of Moody's Corporate Bond Ratings: June 2006 Quarterly Update," Moody's Special Comment, August
2006.
The required adjustments to convert the standard default-based AR measure to a loss-based measure are discussed in "Default & Loss Rates of Structured Finance
Securities: 1993-2003," Moody's Special Comment, September 2004. The concept of accuracy ratio is also described in the Glossary at the end of this report.
See "Default & Loss Rates of Structured Finance Securities: 1993-2005," Moody's Special Comment, April 2006.
See "Structured Finance Rating Transitions: 1983-2006 H1," Moody's Special Comment, August 2006.
Moody's Special Comment
Accuracy Ratios
Figure 3 depicts the one- and five-year accuracy ratios over time with the following notable observations:
• The one-year accuracy ratio for the overall structured finance category continued to trend upward, reaching
95.2% for the latest annual cohort, a remarkable 18.6 percentage points higher than the historical average.
• Both the US RMBS and global CDO sectors experienced dramatic increases in their one-year accuracy
ratios relative to six months ago, jumping 8.3 percentage points to 95.7% for US RMBS and 18.6 percentage points to 93.5% for global CDOs. The US ABS and US CMBS sectors also exhibited good performance, with accuracy ratios of 96.2 % and 89.7% respectively for the most recent annual cohort.
• Conversely, the five-year accuracy ratios for all sectors are still on a declining trend, as the most recent fiveyear cohort ending June 2006 was formed in July 2001 in the middle of the last recession, and significant
improvements in the more recent cohorts have not yet been captured.
Figure 3 - One-Year (solid line) and Five-Year (dotted line) Accuracy Ratios
All Structured Finance
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
US ABS
J
'. ~i"'
~r-\rv~~~~
.
M
en
.".
0>
0>
0>
<D
0>
r-
0>
eo
0>
0>
0>
a
a
:;
N
a
M
a
.".
a
en
a
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0% M
0>
~~
.".
0>
en
0>
<D
0>
r-
0>
M
0>
...
\~
',
.:.
~t;.:.;;·~<
en
0>
<D
0>
r-
0>
eo
0>
0>
0>
a
a
:;
N
a
M
a
.".
a
en
a
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
M
0>
J
en
0>
<D
0>
r-
0>
eo
0>
*
0>
0>
a
a
:;
N
a
M
a
.".
a
en
a
~ ~ ~ ~ ~ ~ '" ~ ~ ~ ~ ~ ~
0:
~
:;
N
a
M
a
en
.".
a
a
M
0>
.".
0>
en
0>
<D
0>
r-
0>
eo
0>
0>
0>
a
a
:;
N
a
M
a
en
.".
a
a
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Int'I SF ex. CDOs
t~
.".
0>
a
a
.
-1A~
,;.,. ··N. I
Global CDOs
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
0>
0>
US RMBS
~~
.".
0>
0>
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
US CMBS
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
eo
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
M
0>
""
~
J
x
~.
.".
0>
en
0>
<D
0>
r-
0>
eo
0>
0>
0>
a
a
:;
N
a
M
a
.".
a
en
a
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Note: At the beginning of each month, securities are grouped together by their alpha-numeric ratings to form rating cohorts. The latest one- and fiveyear cohorts are formed on July 1, 2005 and July 1, 2001. Breaks in the accuracy ratio series occur when the number of impairments for the cohort is
less than or equal to one. Crosses in the one-year series and plus signs in the five-year series indicate that the accuracy ratio was computed from only
one impaired security.
Moody's Special Comment
5
Investment-Grade Loss Rates
Figure 4 shows one-year and five-year investment-grade loss rates by sector. Note that:
• For global structured finance, the one-year investment-grade loss rate was a negligible 0.005% for the most
recent annual cohort, compared to 0.02% and 0.12% six and twelve months prior, respectively.
• For the cohort ending June 2006, the one-year investment-grade loss rate was zero for both US CMBS and
the international structured finance sector excluding CDOs, and has been zero for these two sectors for at
least the last nine months. US RMBS also experienced a zero percent investment-grade loss rate for its four
most recent annual cohorts.
• The one-year investment-grade loss rates fell significantly for US ABS and global CDOs, dropping to
0.005% in the latest annual cohort versus 0.01 % six months ago for US ABS and to 0.02% versus 0.09%
over the same period for global CDOs.
• The five-year investment-grade loss rates in the latest cohorts remained at their historical highs across all
sectors except US RMBS. The five-year loss rate is expected to drop once the more recent cohorts reach
five years of seasoning as the rate has not yet captured the more recent improvements. Additionally, the
five-year investment-grade loss rate in the US RMBS sector was under 0.01 % and the lowest among all sectors since April 2000.
Figure 4 - One-Year (solid line) and Five-Year (dotted line) Investment-Grade loss Rates
All Structured Finance
USABS
2.5%
35%
30%
2.0%
25%
1.5%
20%
15%
1.0%
10%
0.5%
05%
0.0%
:;:
~
g:
<'h
~
~
ill
~
b;
~
8l
~
g:
~
00%
8
~
5
~
2l
~
8
~
"~
~
~
:;:
<'h
~
~
g:
~
ill
b;
8l
~
~
~
USCMBS
g:
~
8
5
2l
8
~
~
~
~
"
~
~
~
USRMBS
20%
18%
18%
16%
16%
14%
14%
12%
12%
10%
10%
08%
08%
i
06%
"\
04%
04%
:;:
<'h
~
~
\
02%
I
02%
00%
06%
~
g:
~
ill
b;
8l
~
~
~
g:
~
00%
8
5
2l
8
~
~
~
~
"
~
~
:;:
<'h
~
~
~
g:
~
ill
b;
8l
~
~
~
~
8
5
2l
8
~
~
~
~
8
5
2l
8
~
~
~
~
"
~
~
~
Int'l SF ex. CDOs
Global CDOs
90%
10%
80%
09%
08%
70%
07%
60%
06%
50%
05%
40%
04%
30%
03%
20%
02%
10%
00%
g:
01%
:;:
<'h
~
~
g:
~
ill
b;
8l
~
~
~
g:
~
00%
8
5
2l
8
~
~
~
~
"
~
~
:;:
<'h
~
~
~
g:
~
ill
b;
8l
~
~
~
g:
~
"
~
~
~
Note: At the beginning of each month. all securities carrying an investment-grade rating are grouped together to form a rating cohort. The latest oneand five-year cohorts are formed on July 1. 2005 and July 1. 2001.
6
Moody's Special Comment
Average Rating Before Impairment
Figure 5 presents the 36-month-average rating before impairment over time as well as the number of newly impaired
securities used to calculate the average ratings. The following observations are noteworthy:
• The average rating before impairment for global structured finance dropped to B2 for the most recent
cohort, two notches down from Ba3 six months prior and a remarkable four notches lower than the historical average. Furthermore, almost all sectors of structured finance experienced declines in their average ratings before impairment.
• The US ABS and global CDO sectors exhibited the sharpest decreases. The US ABS average rating before
impairment was Bl for the most recent cohort versus Ba2 six months prior, while for global CDOs, the
average rating fell to a record low ofB3 versus Baa3 six months ago.
• The average rating before impairment in the US CMBS sector was B3 for the latest annual cohort, the
same as its six months-prior level and one notch lower than its historical average. The average rating before
impairment for US RMBS for the most recent cohort was Ba2, one notch higher than its six month-prior
level and in line with its historical average.
Figure 5 - 36-Month-Average Ratings before Impairment (solid line) and
Number of Newly Impaired Securities (dotted line)
All Structured Finance
USABS
Baa1
500
450
400
Baa2
Baa3
~ ~;
Ba1
/\ rtY\·.
': ,.' '. ~~~
m;J
Ba2
Ba3
B1
;~
\,A .............
.
B2
-·-·T~'-·.
B3
M
en
"'=t"
en
L{)
en
. '.' 100
50
"~"'--c'~~~~~~~~~~~~--,1 0
(,0
en
r-
en
eo
en
en
en
a
a
5
N
a
M
a
""a
Baa1
300
Baa2
250
Baa3
200
Ba1
Ba2
150
Ba3
100
B1
50
0
en
a
M
en
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
"'=t"
en
L{)
en
(,0
en
~ ~ ~ ~
r-
en
"=ro
~
eo
en
Baa2
Baa3
Ba1
~
~
M
en
~
en
L{)
en
~.~~.(~~~
(,0
en
~
en
00
en
en
en
a
a
5
N
a
M
a
""a
en
a
US RMBS
100
90
80
70
60
clj .........
a
a
~ ~ ~ ~ ~ ~ ~ ~
US CMBS
Baa1
Ba3
en
en
~
a
N
a
M
a
~
a
L{)
a
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Baa1
100
90
Baa2
80
70
60
50
40
30
20
10
Baa3
Ba1
Ba2
Ba3
B1
B2
B3
-'-'T~''---'r'~~~~~~~~~~~~~~0
~
~
6;
gs;
g;
25
0
N
a
M
a
~ ~ ~ ~ ~ ~ ~ ~ ~
Moody's Special Comment
7
Figure 5 - 36-Month-Average Ratings before Impairment (solid line) and
Number of Newly Impaired Securities (dotted line)
Global CDOs
Int'I SF ex. CDOs
A3
Baa1
300
250
Baa2
A3
Baa1
10
Baa2
Baa3
200
Baa3
Ba1
Ba2
150
Ba1
Ba2
Ba3
100
Ba3
B1
50
B2
B3
'-'-~~'-'~'~-'~'~'~'~-T"~~~~~~~~~r-!'
M"'=t"L{)(.Or-CO
en
en
en
en
en
en
en
en
a
a
5
N
a
M
a
a
'"
en
0
a
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
"
r
B1
'2
B2
B3 '-'~'T"-'-'~'~'-"~'T'-'-'~"~'-'_'~'-'-"~'~'_':~~'•.~.~.~
•• ~'._.~
M
en
~
en
L{)
en
(,0
en
r-
en
co
en
en
en
a
a
~
a
N
a
M
a
~
a
0
L{)
a
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Note: At the beginning of each month, all securities that become impaired within the next 12-month period are grouped to form a cohort. The latest 12month cohort is formed on July 1, 2005. Breaks in the average rating before impairment series occur when the number of impairments for the cohort
is less than or equal to one. Crosses indicate that the average rating was computed from only one impaired security
Rating Action Rates and Large Rating Action Rates
Figure 6 reports annual rating action rates and large rating action rates. Figures 7 and 8 further disaggregate rating
actions into downgrades and upgrades, and Figure 9 demonstrates how frequently downgrades and upgrades have
been preceded by watchlist actions in the same direction. 7 Key observations include:
• The annual rating action rate in the all structured finance category decreased to its lowest level in almost
four years to 6.9% for the cohort ending June 2006 from 8.0% six months prior and 9.9% twelve months
prior. The large rating action rate also fell to a four-year low of 2.8% in mid-2006 from 3.2% six months
prior and 4.5% a year earlier. The decrease in the rate of rating changes was caused by declines in the frequencies of both downgrades and upgrades over the last year.
• The global structured finance downgrade rate declined to 1.5%, its lowest level since mid-1997 and a moderate decrease from the rate of 2.0% six months earlier, and a marked decrease from 3.3 % one year prior.
The large downgrade rate also remained at low levels at 0.8% in the annual cohort ending June 2006, a
marginal decrease from the rate of 0.9% six months prior, but less than half the year-prior rate of 1.9%. In
addition, the pattern of declining downgrade rates was observed in all sectors of structured finance.
• The overall upgrade rate declined to 5.3 %, lower than the upgrade rate of 6.0% six months ago and the rate
of 6.6% 12 months ago, but still higher than the historical average of 4.1 %. Meanwhile, the large rating
upgrade rate was 2.0% in mid-2006, which was lower than the rates six and twelve months prior, but still
higher than the historical average of 1.5%. The upgrade rate was flat or declining for all sectors of structured finance except for US CMBS where the annual frequency of upgrades jumped to a new high of 19.2 %
inJune 2006.
• The large majority (76.0%) of structured finance downgrades that occurred in the latest annual cohort were
on review for downgrade prior to being downgraded. While this percentage is lower than the proportion of
reviewed downgrades six months prior, it is higher than the historical average. In contrast, only 40.4% of
upgrades in the 12 months preceding June 2006 were on review for upgrade prior to being upgraded, but
this percentage is still greater than the historical average.
7.
8
For the latest rating transition statistics of structured finance securities worldwide, see "Structured Finance Rating Transitions: 1983-2006 H1," Moody's Special Comment, August 2006. Moody's also reviewed its uses of watchlist actions on structured finance securities in a Special Comment, "Structured Finance Watchlist Resolutions: 1992-2003," June 2004.
Moody's Special Comment
Figure 6 - Annual Rating Action Rates (solid line) and Annual large (three notches or more)
Rating Action Rates (dotted line)
USABS
All Structured Finance
14%
14%
12%
12%
10%
10%
8%
8%
6%
6%
4%
4%
2%
2%
0%
~
0;;
~
~
~
~
~
s:
a
~
~
~
~
~
~
~
~
~
~
~
s:
~
0%
~
~
~
1?
Sl
M
m
m
m
m
m
~
~
~
~
~
~
~
m
m
m
a
a
a
~
~
~
~
00
~
~
N
M
a
a
a
~
~
~
~
a
US RMBS
US CMBS
25%
16%
14%
20%
12%
~
10%
15%
8%
10%
6%
. '.
4%
5%
2%
0%
~
0;;
~
~
~
~
~
s:
a
~
~
~
~
~
~
~
~
~
~
~
s:
~
0%
",
1?
Sl
M
m
m
m
m
~
S;
00
m
m
m
a
a
;;
N
M
a
a
a
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
a
~
~
Int'l SF ex. COOs
Global COOs
30%
25%
25%
20%
20%
15%
15%
10%
10%
5%
5%
0%
S1
~
~
~
:§i
S;
g:
g:
~
~
~
~
"
~
;;
~
8
~
~
~
'"
~
:g
~
0%
~
0;;
~
~
~
~
~
s:
a
~
s:
~
~
~
~
~
~
~
~
~
~
~
1?
Sl
~
~
Note: Rating actions include upgrades and downgrades, which are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed
each month covering a 12-month period. The latest 12-month cohort is formed on July 1,2005.
Moody's Special Comment
9
Figure 7 - Annual Downgrade Rates (solid line) and Annual large (three notches or more)
Downgrade Rates (dotted line)
USABS
All Structured Finance
12%
10%
9%
10%
8%
7%
8%
6%
6%
5%
4%
4%
3%
2%
2%
1%
0%
S1
0\
~
~
:g
:§i
S;
g:
g:
~
~
~
~
~
"
~
5
~
8
~
~
~
'"
~
:g
0%
~
~
0;;
~
~
~
~
~
s:
a
~
s:
~
~
~
~
~
~
~
~
~
~
~
'"
~
Sl
~
US RMBS
US CMBS
7%
9%
8%
6%
7%
5%
6%
5%
4%
4%
3%
3%
2%
2%
~A
.....
1%
1%
0%
S1
0\
~
~
:g
:§i
S;
g:
g:
~
~
~
~
~
"
~
5
~
8
~
~
~
'"
~
:g
I
0%
-t
~
0;;
~
~
~
~
~
s:
a
~
s:
<$
Sl
~
~
~
~
~
~
~
~
~
~
~
~
~
~
Int'l SF ex. COOs
Global COOs
18%
30%
16%
25%
14%
12%
20%
10%
15%
8%
6%
10%
4%
5%
0%
S1
0\
~
~
:g
:§i
S;
g:
g:
~
~
~
~
~
"
~
5
~
8
~
~
~
'"
~
:g
~
0%
S1
0\
~
~
:g
:§i
S;
~
~
~
~
~
~
5
~
8
~
~
~
'"
~
:g
~
Note: Downgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month covering a 12-month period.
The latest 12-month cohort is formed on July 1,2005.
10
Moody's Special Comment
Figure 8 - Annual Upgrade Rates (solid line) and Annual large (three notches or more)
Upgrade Rates (dotted line)
All Structured Finance
USABS
9%
10%
8%
9%
7%
8%
7%
6%
6%
5%
5%
4%
4%
3%
3%
2%
2%
1%
1%
0%
S1
0\
~
~
:g
:§i
~
~
S;
g:
g:
~
~
~
"
~
5
~
8
~
~
~
'"
~
:g
0%
~
~
0;;
~
~
~
~
~
s:
a
~
s:
~
~
~
~
~
~
~
~
~
~
~
US CMBS
12%
20%
10%
6%
10%
4%
5%
2%
~
~
~
~
~
~
~
~
~
~
·N,
.'
0%
0%
~
~
J\
8%
15%
~
Sl
~
US RMBS
25%
~
C?
~
0;;
~
~
~
~
~
s:
~
~
~
~
~
~
~
~
Global COOs
a
~
s:
~
~
~
5
~
8
~
~
~
C?
Sl
~
~
Int'l SF ex. COOs
14%
8%
12%
7%
6%
10%
5%
8%
4%
6%
4%
_.)
3%
I
I!
2%
0%
~
,_.,.
S1
0\
~
~
:g
:§i
~
~
.
S;
g:
g:
~
~
~
"
~
5
~
8
~
~
~
'"
~
:g
~
2%
1%
0%
S1
0\
~
~
:g
:§i
S;
g:
g:
~
~
~
~
~
"
~
'"
~
:g
~
Note: Upgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month covering a 12-month period. The
latest 12-month cohort is formed on July 1,2005.
Moody's Special Comment
11
Figure 9 - Percentages of Downgrades (solid line) and
Upgrades (dotted line) Preceded by Watchllst Actions in the Same Direction
All Structured Finance
USABS
100%
100%
90%
90%
80%
80%
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
0%
0;;
~
~
~
~
~
s:
a
~
~
~
~
~
~
~
~
~
s:
~
~
0%
~
~
~
C?
Sl
M
m
m
m
m
m
~
~
~
~
~
~
~
US CMBS
100%
90%
90%
80%
80%
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
~
0;;
~
~
~
~
~
s:
a
~
s:
a
a
a
~
~
~
~
~
N
M
a
a
a
~
~
~
~
a
~
~
~
~
~
~
~
~
~
~
~
.'.
,...~
r~
I
I
I
J
0%
i
C?
Sl
M
m
m
m
m
~
S;
00
m
m
m
a
a
;;
N
a
a
a
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
C?
Sl
~
~
~
~
Global COOs
a
M
~
~
Int'l SF ex. COOs
100%
100%
90%
90%
80%
80%
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
0%
m
m
~
US RMBS
100%
0%
m
00
10%
-·r-,-r~T·---'----'------,·-,-----'------------r·-r·
S1
0\
~
~
:g
:§i
S;
g:
g:
~
~
~
~
~
"
~
;;
~
8
~
~
~
'"
~
:g
~
0%
~
0;;
~
~
~
~
~
s:
a
~
s:
~
~
~
~
~
~
~
~
~
~
~
Note: At the beginning of each month, all securities that experienced downgrades (upgrades) within the next 12-month period are grouped together to
form a rating cohort. The latest 12-month cohort is formed on July 1,2005. Gaps in the data indicate that there were no downgrades (upgrades)
during that time period.
12
Moody's Special Comment
MATERIAllMPAIRMENT
Material impairment, or for simplicity, impairment, is a concept adopted in Moody's structured finance default and
loss studies. Material impairment includes uncured payment defaults and securities downgraded to Ca or C, where
payment defaults include shortfalls of interest and losses of principal on individual tranches of structured finance transactions. A security is called "newly impaired" in a given period if it had no outstanding interest shortfalls, no principal
losses, and was not rated Ca or C in the prior period, but experienced at least one of these three credit events in the
given period for the first time.
lOSS-GIVEN-DEFAUlT (lGD)
Loss-given-default, also known as LGD or loss severity rate, is the total amount of lifetime losses as a share of a
tranche's principal balance on a certain reference date. The losses in each payment period are discounted by a discount
rate, which is typically the stated coupon rate on the tranche. There are three types of principal balances used in the
calculation of LGD: the principal balance at origination, at the time of impairment, and at a cohort formation date.
Depending on the reference principal balance (and the reference date), the calculated LGD can be significantly different due to principal amortization and discounting.
Final LGD on impaired securities with a zero outstanding principal balance are typically known. For impaired
securities with positive principal balances or impaired securities with incomplete loss data, their LGD need to be estimated. Detailed discussions on the estimation of those final LGD are provided in several Moody's Special Comments
including, in particular, "Measuring Loss Severity Rates of Defaulted Residential Mortgage Backed Securities,"
Moody's Special Comment, April 2004, and "Default & Loss Rates of Structured Finance Securities: 1993 -2005," Moody's
Special Comment, April 2006.
ACCURACY RATIO (AR)
An accuracy ratio based on the loss experience of structured finance securities (or an accuracy ratio adjusted for lossgiven-default, or LGD) is the ratio of the area between a loss-based cumulative accuracy profile (CAP) curve and the
45-degree line to the maximum possible area above the 45-degree line. The loss-based CAP curve (or a CAP curve
adjusted for LGD) plots, for each rating category, the proportion of the losses of all impaired securities accounted for
by securities with the same or lower rating against the proportion of all securities in the sample population with the
same or lower rating. 8
To calculate accuracy ratios, we use the same data sample that is used in the latest structured finance default and
loss study (1993-2005) supplemented with default and loss data from the first half of 2006. Rating cohorts are formed
for each calendar month over the study period so that all outstanding securities in the sample population are grouped
together by their alpha-numeric ratings at the beginning of the month. For each monthly rating cohort, we determine
the number of securities that became impaired within one or five years of the cohort formation date and their loss
severity rates as a percentage of the cohort date balance. Cumulative shares of securities rank-ordered by rating are
calculated for the universe of all securities and the universe of impaired securities, respectively, and based on this, a
CAP curve is plotted.
The CAP curve adjusted for LGD is also known as a "power curve" because it shows how effective a rating system
is at differentiating between securities that have sustained high losses from securities that have sustained low or no
losses. The metric is defined relative to the distribution of ratings in the population. The accuracy ratio is an effective
way to summarize the CAP curve into a single number. The accuracy ratio is zero if the CAP curve collapses to the
45-degree line, suggesting that all impaired securities are randomly distributed throughout the population without
regard to rating.
Because we use only data observations that span the time horizon under consideration, the latest cohort for a oneyear horizon is formed onJuly 1, 2005, while for a five-year horizon the latest cohort is formed onJuly 1, 2001.
8.
For an illustration of the CAP curve adjusted for LGD. see "Default & Loss Rates of Structured Finance Securities: 1993-2003." Moody's Special Comment. September 2004.
Moody's Special Comment
13
INVESTMENT -GRADE LOSS RATE
The one-year investment-grade (Aaa through Baa3) loss rate is calculated as follows. First, for a given cohort, we calculate the loss severity as a share of the tranche balance at the cohort date for each security that carried an investmentgrade rating at the cohort formation date and became impaired within a 12-month period after the cohort date. We
then take the sum of these loss rates and divide by the total number of investment-grade securities outstanding as of
the cohort formation date. Note that the loss rate is not weighted by the total dollar volume of outstanding securities
in the cohort. The five-year investment-grade loss rate is calculated similarly.
AVERAGE RATING BEFORE IMPAIRMENT
The rating of an impaired security is measured every month for 36 months prior to impairment. These 36 rating measurements are averaged together to create one representative number for each impaired security. For a particular
cohort, the average rating before impairment is the weighted average of these average ratings for each security that
becomes impaired within 12 months after the cohort formation date, where the weight for each security is the loss
severity rate of the tranche as a share of its original balance. This weighting scheme will place greater emphasis on the
avera~e ratings of impaired tranches with higher loss severity rates over impaired tranches with lower loss severity
rates.
RATING ACTION RATE
The rating action rate is defined as the number of securities that experienced a rating change within a year after cohort
formation divided by the total number of securities outstanding at the cohort formation date (the beginning of each
month). Rating changes are measured on the alpha-numeric (or modified) rating scale. Downgrade rates and upgrade
rates are measured similarly based on downgrade rating actions and upgrade rating actions, respectively.
LARGE RATING ACTION RATE
A large ration action is said to occur if a rating action (or cumulative rating actions) cause(s) a security's rating to
change by three or more notches within a year after cohort formation. The large rating action rate is the number of
such securities divided by the total number of securities outstanding at the cohort formation date. Large downgrade
rates and large upgrade rates are measured similarly based on large downgrade rating actions and large upgrade rating
actions, respectively.
PERCENTAGE OF DOWNGRADES (UPGRADES) PRECEDED BY WATCH LIST ACTIONS IN THE SAME
DIRECTION
This metric is defined as the total number of downgraded (upgraded) securities that were placed on the watchlist in the
same direction before they were downgraded (upgraded), divided by the total number of securities that were downgraded (upgraded) within 12 months after the cohort formation date.
ABS
ABS stand for asset-backed securities. This structured finance sector includes securities backed by home equity loans
(HEL) and both traditional asset types such as auto loans, credit card receivables, student loans, and manufactured
housing loans, and non-traditional asset types such as mutual fund fees, tax liens, tobacco settlement payments, and
intellectual property.
HEL
The home equity loan or HEL sector include securities back by subprime (B&C) mortgage loans, home improvement
loans, high loan-to-value (high LTV) loans, home equity lines of credit (HELOes), and closed-end second-lien loans,
as well as net interest margin (NIM) securitizations. It does not include securities backed by Alt-A mortgages, which
are included in the RMBS sector. HEL is part of the ABS sector.
9.
14
We began using LGD rates as weights in computing an average rating before impairment in the full-year 2005 structured finance performance report. Ideally, LGD
rates should be calculated as a percentage of the principal balance outstanding for each month in the 36 months prior to the impairment. and ratings should then be
weighted by these monthly LGD rates. Practically, however, it does not make any material difference in the average rating number whether we use LGD rates as a
share of impairment-date balance. original balance. or monthly principal outstanding. Since the monthly LGD rates are very time consuming to compute due to amortization. we use the LGD rate as a share of original balance as the weight variable.
Moody's Special Comment
CDOS
CDOs stand for collateralized debt obligations. Derivative securities such as structured notes and repackaged securities are not considered to be part of this sector.
CMBS
CMBS stand for commercial mortgage-backed securities.
RMBS
RMBS stand for residential mortgage-backed securities. The large majority of these securities are backed by first-lien
prime mortgages, but some are backed by Alt-A mortgages. In some older vintage RMBS transactions, subprime
mortgages may also be included in the collateral. HEL is not considered to be part of this sector.
U.S. STRUCTURED FINANCE SECURITIES
u.s. structured finance securities are denominated in U.S. dollars and issued in the U.S. market.
INTERNATIONAL STRUCTURED FINANCE EXCLUDING CDOS
This refers to non-CDO structured finance securities that are not denominated in U.S. dollars or not issued in the
U.S. market. The majority of the securities in this sector are from Europe; the rest comes from the Asia Pacific region,
Canada, and Latin America.
Related Research
Special Comments:
Ih~.f~dQ!fD.~~~~.g.f..~!!J:l.~~!~.gJ~)~~~~.~..R~J~~K~:.X~J.~X~.~!:lQ.Q2.. R~P'Q.~,)\1.~y.lQQ.§..e?Z}±§2
Th.~..~~.ctQ!:!D~g~.~.Qf.~.~~D:I!.~g.X~g.~g.~~..R~!~gg?:.M~.g.~X~.~!:lQ.Q2..R~pgr~,§~.p.~.~.!D!?~!:.~.Q.Q2j2.1±§})
Default & Loss Rates of Structured Finance Securities: 1993-2005, April 2006 (97234)
Default & Loss Rates of Structured Finance Securities: 1993-2004, July 2005 (93653)
Default & Loss Rates of U.S . CDOs: 1993-2003, March 2005 (91692)
Default & Loss Rates of Structured Finance Securities: 2004 First Half Update, January 2005 (90843)
Default & Loss Rates of Structured Finance Securities: 1993-2003, September 2004 (88692)
M~!!.~~rinK1Q.~§..~.~y~!i!Y..R~!~§..Qf.R~.f~~J.~~.g..R~.~ig.~m.~~LMQr~g~g~J?~~k~Q.~~.~~!i!i~§,..Ap.r!U.QQ1..(~§.zQ22
.r.~Y!D.~m..Q.~f~.~!!§..~n.g.M!!!~ri.~UfD.p!!i!:!D~m~.gf.V.:.~:..~.~~D:I!.~g..f~~~~~~..~.~~ll!".~!i~§:J.2.2}~.~QQ.~,..R~~.~!D!?~!:.~.QQ~..c~Q2:±Z2
~.~~~X.~g..f~g~~~..R!!.~g.J.'!!!!l§~~.Qg?:J2.~~.~.~Q.Q§'J-!J.,.A~~?t~QQ.~.{2.~2.z7.)
.~~~~r~.g..f!g!!g~.~.R~!~gKT!:!!g§~.~~.Qg§:J..?~}::2:Q.Q2.L.f..~!?!J:l.~.ry.~.Q.QQ.{2.~2J?')
~.~~.~!:~Q..fj.~.~~~~..R!!.~h~KI1.".~~?i!~Q~?:.J2.~}~.~.QQ.1,.J.<:~Q~~rylQQ.? ..en~n2
Structured Finance Rating Transitions: 1983-2003, February 2004 (81239)
Structured Finance Rating Transitions: 1983-2002, Comparisons with Corporate Ratings and Across Sectors,
January 2003 (77291)
European Structured Finance Rating Transitions: 1988-2005, February 2006 (96706)
The Relationship between Par Coupon Spreads and Credit Ratings in US Structured Finance, December 2005
(95494)
Structured Finance Watchlist Resolutions: 1992-2003, Tune 2004 (87305)
Ih~..~~.ctQr!D!!g~.~.QfMQ.QQi.~..~g!'pQr~!~.l?gn.g..R!!~gg~.:l~~.2:Q.QQ.Q~.~r!~rb!..11:p'g.~!:~l.A~~§t~.QQ§..C2~.12.22
M~!!.~~rinK!h~..r.~dQ!fD.~~~~.g.f~Qrpg!!!.!~..~Q~g.R~.!i~g§,..Ap.r!L~.QQ~jzz.2J.§2
.G~.~g.~.!Q..MQg.gi?.R.~.f~.~h.R~.~.~~!~h:..f~Q~~ry..~.QQ.§JJ'p.g.~.~~)..M!!r~h.lQQ.§..(.?@±!2
To access any of these reports, click on the entry above. Note that these references are current as of the date ofpublication ofthis repo1't
and that more recent rep017S may be available. All research may not be available to all clients.
Moody's Special Comment
15
To order reprints of this report (100 copies minimum), please call 1.212.553.1658.
Report Number:99034
Authors
Associate Analyst
Senior Associate
Production Specialist
Jian Hu
julia Tung
Debjani Dutta Roy
Hadas Alexander
Cassina Brooks
© Copyright 2006, Moody's Investors Service, Inc, and/or Its licensors and affiliates including Moody's Assurance Company, Inc, (together, "MOODY'S"), All rights reserved, ALL
INFORMATION CONTAINED HEREIN IS PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR OTHERWISE REPRODUCED, REPACKAGED,
FURTHER TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN WHOLE OR IN PART. IN
ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER. BY ANY PERSON WITHOUT MOODY'S PRIOR WRITTEN CONSENT. All information contained herein IS obtained by
MOODY'S from sources believed by It to be accurate and reliable, Because of the possibility of human or mechanical error as well as other factors, however, such information IS provided "as
IS" without warranty of any kind and MOODY'S, In particular, makes no representation or warranty, express or Implied, as to the accuracy, timeliness, completeness, merchantability or fitness
for any particular purpose of any such Information, Under no circumstances shall MOODY'S have any liability to any person or entity for (a) any loss or damage In whole or In part caused by,
resulting from, or relating to, any error (negligent or otherwise) or other circumstance or contingency within or outside the control of MOODY'S or any of Its directors, officers, employees or
agents In connection with the procurement, collection, compilation, analysIs, Interpretation, communication, publication or delivery of any such information, or (b) any direct, indirect,
special, consequential, compensatory or incidental damages whatsoever (Including without limitation, lost profits), even If MOODY'S IS advised In advance of the possibility of such
damages, resulting from the use of or inability to use, any such information, The credit ratings and financial reporting analysIs observations, If any, constituting part of the information
contained herein are, and must be construed solely as, statements of opinion and not statements of fact or recommendations to purchase, sell or hold any SeCUrities, NO WARRANTY,
EXPRESS OR IMPLIED, AS TO THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR OTHER
OPINION OR INFORMATION IS GIVEN OR MADE BY MOODY'S IN ANY FORM OR MANNER WHATSOEVER. Each rating or other opinion must be weighed solely as one factor In any
Investment decIsion made by or on behalf of any user of the information contained herein, and each such user must accordingly make ItS own study and evaluation of each security and of
each Issuer and guarantor of, and each provider of credit support for, each security that It may consider purchasing, holding or seiling
MOODY'S hereby discloses that most Issuers of debt seCUrities (including corporate and municipal bonds, debentures, notes and commercial paper) and preferred stock rated by
MOODY'S have, prior to assignment of any rating, agreed to pay to MOODY'S for appraisal and rating services rendered by It fees ranging from $1,500 to $2,400,000, Moody's Corporation
(MCO) and ItS wholly-owned credit rating agency subsidiary, Moody's Investors Service (MIS), also maintain policies and procedures to address the Independence of MIS's ratings and rating
processes, Information regarding certain affiliations that may eXist between directors of MCO and rated entities, and between entities who hold ratings from MIS and have also publicly
reported to the SEC an ownership Interest In MCO of more than 5%, IS posted annually on Moody's website at www,moodys,com under the heading "Shareholder Relations - Corporate
Governance - Director and Shareholder Affiliation Policy"
This credit rating opinion has been prepared without taking Into account any of your obJectives, financial situation or needs, You should, before acting on the OpiniOn, consider the
appropriateness of the opinion having regard to your own obJectives, financial situation and needs
Special Comment
May 2007
Contact
Phone
New York
Julia Tung
Jian Hu
Richard Cantor
1.212.553.1653
The Performance of Structured Finance Ratings:
Full-Year 2006 Report
Highlights
This Special Comment updates Moody's structured finance rating performance metrics as of December 2006, both
with respect to rating accuracy and rating stability.
As indicated in Figure 1, the one-year rating accuracy and stability of the most recent rating cohort (ending
December 2006) was lower than that of the cohort formed six months ago, but higher compared to the performance of
the cohort formed twelve months prior and significantly higher compared to the long-term average performance of all
prior cohorts. The highlights of this report are:
• Overall, 111 structured finance securities became impaired in 2006: 55 in US ABS, 35 in US CMBS, 8 in
US RMBS, 12 in global CDOs, and one in the international structured finance sector excluding CDOs. Of
these, 89 were principal impairments (suffered principal losses or were downgraded to Ca or C), while the
remaining 22 were interest impairments (experienced interest shortfalls only).
• The one-year accuracy ratio for the most recent cohort was 93.5%, which is 1.9 percentage points lower
than its six months-prior level, but 2.0 percentage points higher than its twelve months-prior level.
• The five-year accuracy ratio for the most recent five-year cohort, formed in January 2002, was 60.1 % -below its level of 61.1 % six months earlier and its historical average of 68.3 % -- as this statistic has not yet
captured the more recent improvements in performance.
• Almost all newly impaired securities over the past year were rated speculative-grade well in advance of
impairment. As a result, the one-year investment-grade loss rate was a low 0.02%, less than the loss rate for
the cohort ending December 2005.
• The average rating during the three years prior to impairment of all securities that were impaired for the
cohort ending in December 2006 dropped to an all-time low of E2, one notch below the average rating
prior to impairment for the cohort ending six months earlier and three notches below the average for the
cohort ending twelve months earlier.
• The one-year rating action rate rose to 7.8% in the most recent cohort from its six months-prior level of
6.8%, but was in line with the historical average of 8.0%. The increase was fueled by increases in the frequencies of both downgrades and upgrades over the last six months.
• The one-year large rating action rate (rating changes of three notches or more) was 3.4%, somewhat higher
than the six months-prior level of 2 .8% and the twelve months-prior level of3.2 %.
• The one-year accuracy ratios for all individual sectors of structured finance remained well above their historical averages, while investment-grade loss rates and average ratings before impairment were lower than
their long-term averages (Figure 2). US ABS and global CDOs have shown the most improvement relative
to the past. 1
1.
These performance metrics should be interpreted with caution. Some statistics are based on small samples. as the number of impairments in any given year and any
given sector is often small. Individual performance metrics may also be driven by highly correlated collateral performance under/ying multiple securitizations. particular/y within certain sectors. Moreover, variations in rating performance over time may reflect either changes in the quality of the rating process or changes in the environment that make losses more or less difficult to predict or make collateral performance more volatile. thus resulting in either higher rating accuracy and lower rating
stability at one time than another, or vice versa.
&,Wi;1 Moody's Investors Service
'§i~? W Global Credit Research
Figure 1 - Summary of All Structured Finance Rating Performance as of December 20062
Cohort Ending Date
December 2006
June 2006
December 2005
June 2005
Average (1993-Most Recent)
Number of
New
Impairments
over Prior
12 Months
111
86
83
140
82
................Accuracy Measures· .. ··· .. ·· .. ·· .. ·
36·Month·
l-Year
Average
l·Year
5·Year
Investment·
Rating
Accuracy AccurafY
Grade loss
Before
Ratio
Ratio
Rate
Impairment
93.5%
60.1%
0.02%
B2
95.4%
61.1%
0.00%
B1
91.5%
63.9%
0.02%
Ba2
76.7%
66.2%
0.13%
Bal
76.5%
63.3%
0.19%
Bal
..Stability Measures··
1·Year
large
Rating
1· YearRating
Action
Action Rate
Rate
7.3%
3.4%
6.3%
2.8%
3.0%
3.2%
9.9%
4.5%
3.0%
3.9%
Figure 2 - Summary of Structured Finance Rating Performance by Sector as of December 20062
Number of
New
Impairments
over Prior
12 Months
Cohorts Ending December 2006
USABS
USCMBS
US RMBS
Global COOs
Int'l SF ex. COOs
Historical Averages Since 1993
USABS
USCMBS
US RMBS
Global COOs
Inn SF ex. COOs
2.
3.
2
..... ·· .. ·· .. ··-Accuracy Measures.. ·· .. · ..........
36· Month·
1·Year
Average
l·Year
5·Year
InvestmentRating
Grade loss
Before
Accuracy
AccurafY
Ratio
Ratio
Rate
Impairment
..Stability Measures··
1·Year
large
Rating
1· Year Rating
Action
Action Rate
Rate
55
35
8
12
94.1%
96.1%
94.1%
96.6%
99.5%
53.5%
87.5%
87.3%
59.4%
87.9%
0.04%
0.00%
0.01%
0.00%
0.00%
Bl
Caa2
Ba3
B3
Bl
5.6%
24.3%
4.3%
7.4%
5.4%
3.0%
9.6%
1.9%
3.5%
1.4%
43
8
77.7%
91.2%
69.9%
87.9%
Baa3
Caa1
6.8%
13.3%
5
21
0.4
38.8%
74.2%
65.2%
83.6%
61.0%
76.5%
0.33%
0.01%
0.01%
0.43%
Ba2
Ba2
6.5%
11.3%
4.1%
4.5%
3.0%
6.1%
0.01%
Baa3
5.6%
1.8%
A glossary appears at the end of this report. The number of impairments for historical cohorts is subject to revision during each update as payment shortfalls can be
cured and past remittance or trustee reports may be revised. In addition, consistent with Moody's annual default and loss study, Moody's now derives loss rates using
loss-given-default (LGD) from principal impaired securities alone. For more details, see the entry for LGD in the glossary. The historical average of the number of new
impairments over the prior 12 months is calculated as the total number of newly impaired tranches divided by the number of years in the sample period, and has been
rounded to the nearest integer unless rounding results in zero.
The five-year accuracy ratio is a significantly lagged indicator of performance. It is based on the most recent five-year cohort as of the cohort ending date. For example, the most recent five-year cohort was formed in January 2002.
Moody's Special Comment
Table of Contents
Page
Introduction .................................................................................................................................... 4
Accuracy Ratios .............................................................................................................................. 5
Investment-Grade Loss Rates ......................................................................................................... 6
Average Rating Before Impairment ................................................................................................. 7
Rating Action Rates and Large Rating Action Rates ......................................................................... 8
Glossary ....................................................................................................................................... 13
Related Research ......................................................................................................................... 14
Moody's Special Comment
3
Introduction
In a Special Comment published in April 2003, Moody's developed a set of metrics to measure the performance of corporate ratings with respect to the dual objectives of rating accuracy and rating stability.4 Moody's corporate rating
performance report is now updated on a quarterly basis. s Moody's first introduced and examined its structured
finance rating performance metrics in a September 2004 Special Comment "Default & Loss Rates of Structured
Finance Securities: 1993 -2 003," and published these performance metrics in a stand-alone document for the first time
in September 2005. The structured finance rating performance report is now updated on a semi-annual basis.
For both the corporate and structured finance rating performance reports, the basic unit of observation is a
monthly cohort of ratings, i.e. all outstanding ratings at the beginning of a month are recorded and their performance
tracked over different time horizons. In computing rating performance metrics for structured finance, Moody's incorporates both the default and loss severity experience of all structured finance tranches because Moody's structured
finance ratings rank order expected loss rates. In other words, Moody's structured rating performance metrics weigh
those tranches that have become materially impaired but with lower loss severity less than those with higher loss severity.
The most important measure of rating accuracy is the accuracy ratio, which for structured finance measures the
relationship between tranche ratings and their realized loss rates. 6 This metric measures the quality of Moody's ratings as indicators of relative expected credit loss risk.
As discussed in the April 2003 Special Comment, although relative rating accuracy is our primary objective,
Moody's recognizes that many investors are also concerned with the cardinal accuracy of the rating system. In particular, they expect that investment-grade credits should rarely suffer credit losses and impaired securities should normally carry low ratings well in advance of impairment. For this purpose, we regularly track investment-grade loss rates
and the average rating of securities during the 36 months prior to impairment. Both of these measures should be low
if the rating system is accurate in a cardinal sense.
We employ two measures of rating stability (or rating volatility) - the rating action rate (the frequency of rating
changes) and the large rating action rate (the frequency of rating changes of three notches or more) over a twelvemonth period. To examine how watchlist actions are used, we also report the percentages of downgrades and upgrades
preceded by watchlist (review) actions in the same direction.
Accuracy statistics are calculated based on the data from Moody's latest annual structured finance default and loss
study (1993-2006l For the measurement of stability and volatility, we incorporate the most recent rating action data
from Moody's latest structured finance rating transition study. 8 In particular, the data sample of this report includes
all public, 144A, and private tranches with a Moody's long-term global debt rating in the global asset-backed securities
(ABS), commercial and residential mortgage-backed securities (CMBS and RMBS) and collateralized debt obligations
(CDO) sectors. Provisional ratings, credit evaluations, and national scale ratings are not included. Pari passu tranches
are collapsed into a single tranche. 9 In addition, the following tranches are excluded: (1) tranches guaranteed by
financial guarantors, government agencies, or government-sponsored enterprises (GSEs); (2) interest-only (10) or
residual tranches; (3) repackaged securities, structured notes, structured investment vehicles, structured covered bonds,
and other credit derivative securities; and (4) deals that have all their tranche ratings linked to a single corporate or sovereign rating.
4.
5.
6.
7.
8.
9.
4
See "Measuring the Performance of Corporate Bond Ratings," Moody's Special Comment, April 2003.
For the latest performance report, see "The Performance of Moody's Corporate Bond Ratings: December 2006 Quarterly Update," Moody's Special Comment, February 2007.
The required adjustments to convert the standard default-based AR measure to a loss-based measure are discussed in "Default & Loss Rates of Structured Finance
Securities: 1993-2003," Moody's Special Comment, September 2004. The concept of accuracy ratio is also described in the glossary at the end of this report.
See "Default & Loss Rates of Structured Finance Securities: 1993-2006," Moody's Special Comment, April 2007.
See "Structured Finance Rating Transitions: 1983-2006," Moody's Special Comment, January 2007.
More exactly, tranches carrying the same rating from the same deal, regardless of their rating levels, are collapsed into a single rating observation, with the following
exception: if two or more tranches share the same rating in the same deal, but are collateralized by distinct groups of loan pools, then the tranches are not collapsed.
Additionally, we do not review each tranche of every deal in order to determine whether it is pari passu to another tranche of the same deal.
Moody's Special Comment
Accuracy Ratios
Figure 3 depicts the one- and five-year accuracy ratios over time with the following notable observations:
• The one-year accuracy ratio for the overall structured finance category dropped slightly from its level six
months ago from 95.4% to 93.5%, but was still higher than its level 12 months and 24 months ago.
• All individual structured finance sectors exhibited one-year accuracy ratios above 94% for the January 2006
cohort. lO US RMBS made the greatest improvement from its six-month prior level, increasing 3.2 percentage points to 94.1 % over six months, while global CDOs experienced the largest year-over-year
increase, jumping 11.6 percentage points to 96.6% over a twelve-month period. US ABS and CMBS were
basically flat for the year at roughly 95%.
• Conversely, there was little improvement seen in the five-year accuracy ratios for all sectors, with the exception of the international structured finance sector excluding CDOs, as the most recent five-year cohorts
were formed in late 2001 in the midst of the last recession, and significant improvements in the more recent
cohorts have not yet been captured.
Figure 3 - One-Year (solid line) and Five-Year (dotted line) Accuracy Ratios
USABS
All Structured Finance
100%
.. .r.~
90%
80%
~
70%
)I(
,
'"
•.••.•••
60%
100%
90%
80%
70%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
0%
0%
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
Cohort Starting Date
80%
70%
100%
8
~
~
8
~
g
~
~
~
8
~
80%
)I(
40%
~;:::'t;::;~
30%
20%
20%
10%
10%
0%
0%
~
~
~
~
~
~
~
~
~
~
~
~
~
~
8
~
,;
~
8
~
8
~
~
~
8
q;
9
--------------
~
Cohort Starting Date
Cohort Starting Date
Global COOs
100%
~
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
,;
50%
40%
30%
~
60%
,.
50%
8
~
~~~~
90%
70%
60%
~
US RMBS
.~
90%
~
Cohort Starting Date
US CMBS
100%
~
Int'I SF ex. COOs
100%
,-~
90%
I
80%
70%
~
LJ
60%
50%
40%
30%
20%
10%
0%
q;
----------- -Cohort Starting Date
~
~
~
~
~
~
~
~
~
~
~
~
~
~
Cohort Starting Date
Note: At the beginning of each month, securities are grouped together by their alpha-numeric ratings to form rating cohorts. The latest one- and fiveyear cohorts are formed on January 1, 2006 and January 1, 2002. Breaks in the accuracy ratio series occur when the number of impairments for the
cohort is less than or equal to one. Crosses in the one-year series and plus signs in the five-year series indicate that the accuracy ratio was computed
from only one impaired security.
10. The figures for US ABS and RMBS in this study differ from those reported in prior studies because some of the old vintage RMBS transactions sponsored by DLJ
were reclassified into the HEL (and therefore ABS) category. Please refer to "Deal Sponsor and Credit Risk of US ABS and MBS Securities," Moody's Special Comment, December 2006 which discusses how these transactions were backed by subprime mortgage loans by today's standard. Please also see the glossary for the
definition of RMBS, HEL and ABS.
Moody's Special Comment
5
Investment-Grade Loss Rates
Figure 4 shows one-year and five-year investment-grade loss rates by sector. Note that:
• For global structured finance, the one-year investment-grade loss rate rose to 0.02% from 0.004% six
months prior, but was still low both in absolute terms, and relative to the loss rate 12 months and 24 months
pnor.
• For the cohort formed in January 2006, the one-year investment-grade loss rate was zero for US CMBS,
global CDOs, and the international structured finance sector excluding CDOs, and has remained zero for
the last 15, 9, and 12 months in each of these three sectors, respectively.
• The one-year investment-grade loss rate increased from 0.006% to 0.04% for US ABS over a six-month
period, but was far lower than that of most of the cohorts formed before 2005. The US RMBS investmentgrade loss rate for the January 2006 cohort was 0.007%, slightly below the 0.008% loss rate for the July
2005 cohort and much smaller than the January 2005 rate of 0.03%.
• All sectors displayed a declining trend in their five-year investment-grade loss rates, reflecting the improvement in performance for the more recent cohorts. Global CDOs experienced the steepest decline, falling
from 6.1% for the January 2001 cohort to 4.1% for the January 2002 cohort. Additionally, the five-year
investment-grade loss rate for US RMBS remained the lowest among all the sectors at 0.002%.
Figure 4 - One-Year (solid line) and Five-Year (dotted line) Investment-Grade loss Rates
All Structured Finance
USABS
25%
40%
35%
20%
30%
25%
15%
20%
10%
15%
10%
05%
05%
co
Cohort Starting Date
0)
0)
0)
0
0
0
~
~
~
~
~
Cohort Starting Date
US CMBS
US RMBS
20%
025%
18%
16%
. ",
14%
"
020%
12%
015%
10%
08%
"
.
06%
04%
:
•
02%
:
i
,
010%
"I
005%
!
OO%j-..·~~~~~~~~"""';'~...-.,.l~~~~~
~
Cohort Starting Date
~
~
g
~
~
~
~
~
~
~
~
~
~
Cohort Starting Date
Int'l SF ex. COOs
Global COOs
80%
70%
60%
50%
40%
30%
20%
10%
~
~
~
co
~
~
~
0)
0)
0)
0
0
0
~
~
~
~
g
~
Cohort Starting Date
~
~
~
~
~
Cohort Starting Date
Note: At the beginning of each month, all securities carrying an investment-grade rating are grouped together to form a rating cohort. The latest oneand five-year cohorts are formed on January 1,2006 and January 1,2002.
6
Moody's Special Comment
Average Rating Before Impairment
Figure 5 presents the 36-month-average rating before impairment over time as well as the number of newly impaired
securities used to calculate the average ratings. The following observations are noteworthy:
• The average rating before impairment for global structured finance fell to B2 for the most recent cohort,
one notch down from B 1 six months prior and three notches below its level twelve months prior. Moreover, the average ratings before impairment for all sectors of structured finance remained below their historical averages.
• The US ABS and global CDO sectors exhibited the sharpest declines relative to the recent past. The average ratings before impairment for US ABS and global CDOs were Bl and B3, respectively, three and six
notches below their year-prior averages.
• As has been the case in the past, US CMBS exhibited the lowest average rating before impairment of all sectors at Caa2, which is a historical low. The average rating before impairment for US RMBS for the most
recent cohort was Ba3, one notch lower than its level six months ago.
Figure 5 - 36-Month-Average Ratings before Impairment (solid line) and
Number of Newly Impaired Securities (dotted line)
USABS
All Structured Finance
Baa1
500
Baa1
Baa2
450
Baa2
400
Baa3
Baa3
350
300
250
200
150
61
50
o
300
l
150
100
62
0
~~~~~~~
Cohort Starting Date
Cohort Starting Date
US CMBS
US RMBS
Baa2
Baa3
61
62
63
Caa1
Caa2 +-_~C-T_'-'-_"O-_"_-'-'-'_~'->!-~~~~~~--;C
Cohort Starting Date
Cohort Starting Date
Global COOs
Int'I SF ex. COOs
A3
200
A3
Baa1
180
Baa1
Baa2
160
Baa2
Baa3
140
Baa3
6,1
120
6,2
100
6,3
80
61
62
60
63
20
10
61
62
63~~~~~~~~~~~~~~~~~
Cohort Starting Date
Cohort Starting Date
Note: At the beginning of each month. all securities that become impaired within the next 12-month period are grouped to form a cohort. The latest 12month cohort is formed on January 1. 2006. Breaks in the average rating before impairment series occur when the number of impairments for the
cohort is less than or equal to one. Crosses indicate that the average rating was computed from only one impaired security
Moody's Special Comment
7
Rating Action Rates and Large Rating Action Rates
Figure 6 reports 12-month rating action rates and large rating action rates. Figures 7 and 8 further disaggregate rating
actions into downgrades and upgrades, and Figure 9 demonstrates how frequently downgrades and upgrades have
been preceded by watchlist actions in the same direction. 11 Key observations include:
• The 12-month rating action rate in the all structured finance category was 7.8% for the January 2006
cohort, an increase from 6.8% six months prior, but in line with the twelve months-prior and historical levels (both 8.0%). The large rating action rate also increased over the last six months to 3.4% from 2.8%, but
was in line with the twelve months-prior level of 3.2 % and below the historical average of 3.9%.
• The global structured finance downgrade rate increased moderately to 1.9% from 1.5% six months earlier,
but was still far below the peak of 8.8% reached by the May 2002 cohort. The large downgrade rate also
rose slightly from 0.8% to 0.9% over the same period. The increase in the overall downgrade rate was
caused by the rise in negative rating actions among US ABS, including securities backed by home equity
loans and manufactured housing loans, and global CDOs, including resecuritization CDOs, synthetic arbitrage CDOs, and high-yield CBOs.
• The overall upgrade rate for the most recent cohort was 5.8%, higher than the upgrade rate of 5.3% six
months ago, but slightly lower than the rate of 6.0% twelve months ago. Meanwhile, the large rating
upgrade rate was 2.5%, up from 2.0% six months prior and 2.3 % twelve months prior. The upgrade rate
was on an increasing trend for all sectors of structured finance except US RMBS and international structured finance excluding CDOs. In addition, the frequencies of all upgrades and large upgrades for US
CMBS reached all-time highs for the January 2006 cohort of 22.3% and 9.1 %, respectively.
• Roughly three-quarters of structured finance downgrades that occurred over the last twelve months were
on review for downgrade prior to being downgraded. However, the proportion of downgrades reviewed
varied greatly by sector, with US ABS and RMBS exhibiting rates of approximately 90% each and US
CMBS showing a rate of only 11 %. In contrast, a much smaller percentage of upgrades were on review for
upgrade prior to the rating action. The fraction of upgrades in the January 2006 cohort that were preceded
by watchlist actions in the same direction was 29%, down from 41 % in the July 2005 cohort. 12
11. For the latest rating transition statistics of structured finance securities worldwide. see "Structured Finance Rating Transitions: 1983-2006." Moody's Special Comment, January 2007. Moody's also reviewed its uses of watchlist actions on structured finance securities in a Special Comment, "Structured Finance Watchlist Resolutions: 1992-2003," June 2004.
12. In 2006, there were 993 watch list actions for downgrades affecting 913 unique tranches and 903 watch list actions for upgrades affecting 871 unique tranches. 68% of
the reviews for downgrades resulted in an actual downgrade, 17% resulted in confirmations or affirmations, 12% were ongoing as of the end of the first quarter of
2007, and the remaining 2% of the reviews concluded with a rating withdrawal or upgrade. 83% of the reviews for upgrade resulted in an actual upgrade, 9% resulted
in confirmations or affirmations, 4% were ongoing as of the end of 2007 Q1, and the remaining 5% concluded with a rating withdrawal.
8
Moody's Special Comment
Figure 6 - 12-month Rating Action Rates (solid line) and
12-month large (three notches or more) Rating Action Rates (dotted line)
All Structured Finance
USABS
Cohort Starting Date
Cohort Starting Date
US CMBS
US RMBS
16%
14%
12%
oli
10%
8%
6%
4%
2%
. ,J.~.~
O%~~~~~~~~~~~~~~~~~~~
Cohort Starting Date
Cohort Starting Date
Int'l SF ex. COOs
Global COOs
30%
20%
18%
25%
16%
14%
20%
12%
15%
10%
8%
10%
5%
6%
4%
~,
2%
O%~~T·~~·~·~~~,~~T-~~~~~~~~~~~
0%
Cohort Starting Date
Cohort Starting Date
Note: Rating actions include upgrades and downgrades, which are measured on the alpha~numeric (or modified) rating scale. Rating cohorts are
formed each month covering a 12~month period. The latest 12~month cohort is formed on January 1,2006.
Moody's Special Comment
9
Figure 7 -12-month Downgrade Rates (solid line) and
12-month large (three notches or more) Downgrade Rates (dotted line)
USABS
All Structured Finance
10%
12%
9%
10%
8%
7%
8%
6%
5%
6%
4%
4%
3%
2%
2%
1%
0%
0%
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
Cohort Starting Date
~
8
~
,;
~
8
~
8
~
g
~
~
~
8
~
Cohort Starting Date
US CMBS
9%
~
US RMBS
8%
8%
7%
6%
5%
~
7%
6%
5%
4%
, ..........
4%
3%
2%
1%
0%
~
~
~
~
~
~
~
~
3%
2%
1%
0%
~
~-~
q;
~
~
~
9
~
~
~
Cohort Starting Date
~
~
~
~
~
~
~
~
Cohort Starting Date
Int'l SF ex. COOs
Global COOs
16%
30%
14%
25%
12%
20%
10%
15%
8%
6%
10%
4%
5%
2%
0%
0%
q;
~
8
~
~
Cohort Starting Date
~
~
~
~
~
~
~
~
~
~
~
~
g
~
~
~
8
~
8
~
3
~
g
~
Cohort Starting Date
Note: Downgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month covering a 12-month period.
The latest 12-month cohort is formed on January 1, 2006.
10
Moody's Special Comment
Figure 8 - 12-month Upgrade Rates (solid line) and
12-month large (three notches or more) Upgrade Rates (dotted line)
USABS
All Structured Finance
9%
9%
8%
8%
7%
7%
6%
6%
5%
5%
4%
4%
3%
3%
2%
2%
1%
1%
0%
~
~
~
~
~
~
~
~
~
~
~
~
8
~
,;
~
8
~
8
~
g
~
0%
~
~
8
8
~
~
~
~
~
~
~
~
~
~
~
~
g
~
~
~
~
~
8
~
8
~
~
3
~
~
g
~
Cohort Starting Date
Cohort Starting Date
US RMBS
US CMBS
25%
12%
J\
10%
20%
8%
15%
6%
10%
'.
4%
5%
2%
0%
0%
~
~
~
~
~
~
~
~
~
~
~
~
~
g
~
'
...
'N
~
~
8
~
8
~
3
~
~
~
g
~
Cohort Starting Date
Cohort Starting Date
Int'l SF ex. COOs
Global COOs
8%
14%
7%
12%
6%
10%
5%
8%
4%
6%
3%
4%
2%
2%
1%
0%
8
~
~
~
~
~
~
~
~
~
~
~
~
~
g
~
~
~
Cohort Starting Date
8
~
8
~
3
~
~
~
g
~
0%
8
~
~
~
~
~
Cohort Starting Date
Note: Upgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month covering a 12-month period.
The latest 12-month cohort is formed on January 1, 2006.
Moody's Special Comment
11
Figure 9 - Percentages of Downgrades (solid line) and
Upgrades (dotted line) Preceded by Watchllst Actions in the Same Direction
USABS
All Structured Finance
100%
100%
90%
90%
80%
80%
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
0%
0%
~
~
~
Cohort Starting Date
Cohort Starting Date
US CMBS
US RMBS
100%
100%
90%
90%
80%
80%
70%
60%
50%
40%
30%
20%
~
. ..
10%
0%
'.
",
q;
~
~
~
~
~
~
~
~
~
~
'.
~
',-','-"---'-
~
rY~
"
70%
'
I:'
60%
"
I
50%
40%
30%
20%
10%
0%
~
~
Cohort Starting Date
~
~
~
Cohort Starting Date
Int'I SF ex, COOs
Global COOs
100%
100%
90%
~
80%
70%
60%
50%
~"'"
."'
40%
90%
80%
70%
60%
50%
40%
30%
30%
20%
20%
10%
10%
0%
0%
~
~
~
~
~
~
~
~
~
Cohort Starting Date
~
~
~
.' ,
~
Cohort Starting Date
Note: At the beginning of each month, all securities that experienced downgrades (upgrades) within the next 12-month period are grouped together to
form a rating cohort. The latest 12-month cohort is formed on January 1, 2006, Gaps in the data indicate that there were no downgrades (upgrades)
during that time period,
12
Moody's Special Comment
PAYMENT SHORTFALL
Structured finance securities are defined as being in payment shortfall (previously called "payment default") if they
have suffered:
•
•
an interest shortfall, or
a principal write-down.
MATERIALIMPAIRMENT
Structured finance securities are defined as being in material impairment if they have:
• sustained a payment shortfall that remained uncured, or
• been downgraded to Ca or C.
Prepayment-related and AFC-related interest shortfalls are not considered to be material impairments, but PIKing tranches are. Explicit principal write-downs are included whereas implicit principal write-downs or under-collateralizations are not.
The impairment status of a security may change as it goes from cured (i.e. all outstanding shortfalls and losses
were repaid in full) to uncured (i.e. positive interest shortfalls or principal losses outstanding), or vice versa. If any securities rated Ca or C but not in payment shortfall are upgraded, they are considered to be no longer in material impairment. Securities rated Ca or C that were not upgraded are in material impairment even if their payment shortfalls have
been cured. Finally, securities with very minor shortfalls or losses are excluded.
A security is called "newly impaired" in a given period if it had no payment shortfalls and was not downgraded to
Ca or C in the prior period, but experienced at least one of these two credit events in the given period for the first time.
PRINCIPAL IMPAIRMENT
This refers to materially impaired securities that have suffered principal write-downs or principal losses, or have been
downgraded to Ca or C even though a principal write-down or loss has not yet been observed. In particular, if a security had experienced principal write-downlloss or was downgraded to Ca or C, it is called a principal impairment
regardless of whether it had experienced interest shortfalls.
INTEREST IMPAIRMENT
This refers to materially impaired securities that have experienced only interest shortfalls, no principal losses, and were
not downgraded to Ca or C.
INVESTMENT -GRADE (lG) SECURITIES
This refers to securities rated Aaa, Aal, Aa2, Aa3, AI, A2, A3, Baal, Baa2, or Baa3.
SPECULATIVE-GRADE (SG) SECURITIES
This refers to securities rated Bal, Ba2, Ba3, BI, B2, B3, Caal, Caa2, Caa3, Ca, or C.
LOSS-GIVEN-DEFAULT (LGD) OR LOSS SEVERITY RATE
The loss-given-default, also known as LGD or loss severity rate, of an impaired security is measured by the sum of the
present values of net losses, including both interest shortfalls and principal losses, discounted by the security's coupon
rate and expressed as a percentage of the principal balance on a certain reference date. There are typically three types
of reference dates used in the calculation ofLGD: origination, the time of impairment, and a cohort formation date.
Final LGD on impaired securities with a zero outstanding principal balance are typically known. For impaired
securities with positive principal balances or impaired securities with incomplete loss data, their LGD need to be estimated. 13 In addition, consistent with Moody's annual default and loss study, Moody's now derives loss rates using
LGD from principal impaired securities alone. In almost all cases, securities with only interest shortfalls (interest
impairments) end up being cured or suffering principal writedowns or being downgraded to CalC (principal impairments). Interest impairments have frequently been cured, whereas cures on principal impairments have been rare. As
a result, we do not attempt to project final LGD on interest impaired securities until they either become cured, in
which case they drop out of the LGD data sample with an LGD of zero, or become principal impaired, in which case a
projection of the final LGD is made if the impairment is not resolved at that time.
13. See "Default & Loss Rates of Structured Finance Securities: 1993-2006." Moody's Special Comment, April 2007 and "Measuring Loss-Given-Default for Structured
Finance Securities: An Update," Moody's Special Comment, December 2006 for discussions of the characteristics and estimation methods of LGD.
Moody's Special Comment
13
ACCURACY RATIO (AR)
An accuracy ratio based on the loss experience of structured finance securities (or an accuracy ratio adjusted for lossgiven-default, or LGD) is the ratio of the area between a loss-based cumulative accuracy profile (CAP) curve and the
45-degree line to the maximum possible area above the 45-degree line. The loss-based CAP curve (or a CAP curve
adjusted for LGD) plots, for each rating category, the proportion of the losses of all impaired securities accounted for
by securities with the same or lower rating against the proportion of all securities in the sample population with the
same or lower rating. 14
To calculate accuracy ratios, we use the same data sample that is used in the latest structured finance default and
loss study (1993-2006). Rating cohorts are formed for each calendar month over the study period so that all outstanding securities in the sample population are grouped together by their alpha-numeric ratings at the beginning of the
month. For each monthly rating cohort, we determine the number of securities that became impaired within one or
five years of the cohort formation date and their loss severity rates as a percentage of the cohort date balance. Cumulative shares of securities rank-ordered by rating are calculated for the universe of all securities and the universe of
impaired securities, respectively, and based on this, a CAP curve is plotted. Note that only LGD for principal
impaired securities are used in the calculation. Please see the definition ofLGD for further details.
The CAP curve adjusted for LGD is also known as a "power curve" because it shows how effective a rating system
is at differentiating between securities that have sustained high losses from securities that have sustained low or no
losses. The metric is defined relative to the distribution of ratings in the population. The accuracy ratio is an effective
way to summarize the CAP curve into a single number. The accuracy ratio is zero if the CAP curve collapses to the
45-degree line, suggesting that all impaired securities are randomly distributed throughout the population without
regard to rating.
Because we use only data observations that span the time horizon under consideration, the latest cohort for a oneyear horizon is formed onJanuary 1, 2006, while for a five-year horizon the latest cohort is formed onJanuary 1, 2002.
INVESTMENT -GRADE lOSS RATE
The one-year investment-grade loss rate is calculated as follows. First, for a given cohort, we calculate the LGD as a
share of the tranche balance at the cohort date for each security that carried an investment-grade rating at the cohort
formation date and became impaired within a 12-month period after the cohort date. We then take the sum of these
LGD rates and divide by the total number of investment-grade securities outstanding as of the cohort formation date.
Note that the LGD rate is not weighted by the total dollar volume of outstanding securities in the cohort. Also note
that only LGD for principal impaired securities are used in the calculation. Please see the definition of LGD for further details. The five-year investment-grade loss rate is calculated similarly.
AVERAGE RATING BEFORE IMPAIRMENT
The rating of an impaired security is measured every month for 36 months prior to impairment. These 36 rating measurements are averaged together to create one representative number for each impaired security. For a particular
cohort, the average rating before impairment is the weighted average of these average ratings for each security that
became impaired within 12 months after the cohort formation date, where the weight for each security is the LGD
rate of the tranche as a share of its original balance. This weighting scheme will place greater emphasis on the average
ratings of impaired tranches with higher LGD over impaired tranches with lower LGD. 15 Note that only LGD for
principal impaired securities are used in the calculation. Please see the definition ofLGD for further details.
RATING ACTION RATE
The rating action rate is defined as the number of securities that experienced a rating change within a year after cohort
formation divided by the total number of securities outstanding at the cohort formation date (the beginning of each
month). Rating changes are measured on the alpha-numeric (or modified) rating scale. Downgrade rates and upgrade
rates are measured similarly based on downgrade rating actions and upgrade rating actions, respectively.
14. For an illustration of the CAP curve adjusted for LGD. see "Defauff & Loss Rates of Structured Finance Securities: 1993-2003." Moody's Special Comment, September 2004.
15. We began using LGD rates as weights in computing an average rating before impairment in the full-year 2005 structured finance performance report. Ideally, LGD
rates should be calculated as a percentage of the principal balance outstanding for each month in the 36 months prior to the impairment, and ratings should then be
weighted by these monthly LGD rates. Practically, however, it does not make any material difference in the average rating number whether we use LGD rates as a
share of impairment-date balance, original balance, or monthly principal outstanding. Since the monthly LGD rates are very time consuming to compute due to amortization, we use the LGD rate as a share of original balance as the weight variable.
14
Moody's Special Comment
LARGE RATING ACTION RATE
A large ration action is said to occur if a rating action (or cumulative rating actions) cause(s) a security's rating to
change by three or more notches within a year after cohort formation. The large rating action rate is the number of
such securities divided by the total number of securities outstanding at the cohort formation date. Large downgrade
rates and large upgrade rates are measured similarly based on large downgrade rating actions and large upgrade rating
actions, respectively.
PERCENTAGE OF DOWNGRADES (UPGRADES) PRECEDED BY WATCH LIST ACTIONS IN THE SAME DIRECTION
This metric is defined as the total number of downgraded (upgraded) securities that were placed on the watchlist in the
same direction before they were downgraded (upgraded), divided by the total number of securities that were downgraded (upgraded) within 12 months after the cohort formation date.
ABS
ABS stand for asset-backed securities. This structured finance sector includes securities backed by home equity loans
(HEL) and both traditional asset types such as auto loans, credit card receivables, student loans, and manufactured
housing loans, and non-traditional asset types such as mutual fund fees, tax liens, tobacco settlement payments, and
intellectual property.
HEL
The home equity loan or HEL sector includes securities backed by subprime (B&C) mortgage loans, home improvement loans, high loan-to-value (high LTV) loans, home equity lines of credit (HELOCs), and closed-end second-lien
loans, as well as net interest margin (NIM) securitizations. It does not include securities backed by Alt-A mortgages,
which are included in the RMBS sector. HEL is part of the ABS sector.
Prior to 1998, RMBS collateral was generally defined as first-lien residential mortgages, regardless of the credit quality of the borrower. HEL collateral generally included junior liens such as HELOCs or closed-end seconds. However, as
subprime lending became more prevalent, the market shifted its definition such that HEL encompassed subprime firstlien residential mortgages while RMBS included first-lien mortgages made to higher quality borrowers. Since 1998 and
especially in the last five years, a deal classified as RMBS by Moody's is generally backed by prime or Alt-A quality firstlien residential mortgages, while a deal classified as HEL can be backed by subprime first-lien mortgages or junior liens.
Therefore, a subprime deal which would be classified as HEL today may have been classified as RMBS in the past.
CDOS
CDOs stand for collateralized debt obligations. Derivative securities such as structured notes and repackaged securities are not considered to be part of this sector. Commercial real estate (CRE) CDOs, where 70% or more of the collateral is comprised of CRE loans, are classified as CMBS. If the collateral backing the transaction contains less than
70% CRE loans, then the deal is classified as a CDO.
CMBS
CMBS stand for commercial mortgage-backed securities. Commercial real estate (CRE) CDOs, where 70% or more
of the collateral is comprised of CRE loans, are classified as CMBS. If the collateral backing the transaction contains
less than 70% CRE loans, then the deal is classified as a CDO.
RMBS
RMBS stand for residential mortgage-backed securities. The large majority of these securities are backed by first-lien
prime mortgages, but some are backed by Alt-A mortgages. For further details, see the definition ofHEL.
ALL STRUCTURED FINANCE SECURITIES
All structured finance captures global structured finance securities in four major sectors: ABS, CDO, CMBS, and RMBS.
U.S. STRUCTURED FINANCE
u.S. structured finance securities are denominated in U.S. dollars and issued in the U.S. market or denominated in
Canadian dollars and issued in Canada. In cases where the source of the underlying collateral and the denomination of
the securities crossed multiple countries/regions, deals are classified by the location at which they are monitored.
INTERNATIONAL STRUCTURED FINANCE EXCLUDING CDOS
This refers to non-CDO structured finance securities that are not denominated in U.S. dollars and issued in the U.S.
market and not denominated in Canadian dollars and issued in Canada. The majority of the securities in this sector are
issued in Europe, the Middle East, and Africa (EMEA); the rest are issued in the Asia Pacific region and Latin America.
Moody's Special Comment
15
Related Research
Ih_~_~~_ct9gg~g~~_gf~_~~!gx_~gJ~Ig~g~~J~~!!gg~:_.M!Q~X~~£_~_QQ§J~~.RQ!!)__~_~PJ~~!?_~£_~_QQ§__(22QH)
The Performance of Structured Finance Ratings: Full-Year 2005 Report, May 2006 (97346)
:rh~J~~rt:Q£~~DS_~_9L~!!J!_~m£~QJ:\~~~DS_~_R~!igg~:_MJ_g::Y~~L~QQIR~pgnL~~.R!~~Q~t~_Q_Q~__(211QJ2
Default & Loss Rates of Structured Finance Securities: 1993-2006, April 2007 (102733)
R~f~~!!_~__l-:9_~_~__R~!~~_Qf~_~!J:Ic~!g!:~_~}Ig_~g_~~ __~~S_~!:j_~_~?_:J22J_:lQ_Q~_L!.\.RrtLlQ_Q§_J2t~}12
Default & Loss Rates of Structured Finance Securities: 1993-2004, July 2005 (93653)
P~f~_~lt~J:_Q~~_R~~_~_~_gfJl·_~_·__ ~PQ~:)9._2}~lQQ}L~I~x_~hJQQ_?j2J§2~2
Default & Loss Rates of Structured Finance Securities: 2004 First Half Update, January 2005 (90843)
R~f~~!!.~..l-:9.~.~..R~!~~.Qf.~.~!J:Ic~!g!:~.~..f.~g.~g.~~ ..~~S.~!:j.~.~?:J22J.:.~.Q.Q}t..~~J2!~~Q~!:)Q.Q1.C~~.Qn2
M~~~~r!gg_l-:9_~_~_:~;hY~D::R~f~~!!JQX__~~~!g£~_~X!g_~_~~_~~S~£!!!_~~_:_~_l!.RQ~!~,J2~s~_~!?_~r_~_QQ_QJtQE_~12
Measuring Loss Severity Rates of Defaulted Residential Mortgage Backed Securities: AMethodology, April 2004 (86769)
~_~~~mJ2_~f~~I~_~Q_M~!~rt_~V~J2~!X~_~g_t~_Qf1!.:~:__~!!J!_<::m£~QX!D_~S_~_~~s~rt!!~_~:_J22J::~_Q_Q~.LR~_<::~~Q~£J_Q_QLC~Q_~122
Structured Finance Rating Transitions: 1983-2006,January 2007 (101833)
~_~~!gx_~gJ~Ig~~~__R~!iggT£~g~!!i9D_~:J2~}_::~_QQ§J::U.L~~@~t~_QQ§_J2~_EZ2
Structured Finance Rating Transitions: 1983-2005, February 2006 (96533)
.~!!J!_~_m£~QX~_~S_~_R~~DgTJ;_~_~!_~_Qg~:J9._~J_~_~QQ1Lf_~!?!:~~~rylQQ_?j9.J}2l)
Structured Finance Rating Transitions: 1983-2003, February 2004 (81239)
~.~~!gt~(f~~~~..R~!!gg.JJ~~~!!9.g.~:J.2~}::~.Q.Q~).l~~.~.ry)Q.QH.?Z.n.D
European Structured Finance Rating Transitions: 1988-2006, February 2007 (10203 3)
~~rQP~~g__~!!J!s_m£~QX~~S_~__R~_~DgTJ;_~D~!_~_Qg~:J9._~~_:lQ_Q~_lX~!?~~ryJQQ_Q __(9._QtQ§2
Japanese Structured Finance Rating Transitions: 1994-2006, March 2007 (102509)
Asia Pacific (ex-Japan) Structured Finance Rating Transitions: 1990-2006 (102095)
R~_~L~pgg~9L~g__~X_~g!J_R!~kgUL~_:_A1?_~__~g__M!?_~__~_~~~rt!!~~,J2~s_~~!?_~£_~_QQ§JtQQ~n2
The Relationship between Par Coupon Spreads and Credit Ratings in US Structured Finance, December 2005 (95494)
_~!!J!s_m£~QX~~s_~_1Y~!~hE_gJ~~~91~_~gg~:)9._2~_~_~QQ}_lJ~g~_~_Q_Q1J~2_JQ~2
The Performance of Moody's Corporate Bond Ratings: December 2006 Quarterly Update, February 2007 (101930)
M~~~~r!gg_th~__r_~rt:Q~~S_~_9L~QmQ!:~!~__~_QDQ_R~!!gg~L~PX_!UQ_QLQZ21§2
Guide to Moody's Default Research: April 2007 Update, April 2007 (102661)
To access any of these reports, click on the entry above. Note that these references are current as of the date ofpublication ofthis repo1't
and that more recent rep017S may be available. All research may not be available to all clients.
To order reprints of this report (100 copies minimum), please call 1.212.553.1658.
Report Number: 103017
Authors
Associate Analyst
Senior Associate
Senior Production Associate
Jian Hu
julia Tung
Debjani Dutta Roy
Hadas Alexander
Charles Ornegri
© Copyright 2007, MOOdy's Investors SerVice, Inc, and/or Its licensors and affiliates including Moody's Assurance Company, Inc, (together, "MOODY'S"), All rights reserved, ALL
INFORMATION CONTAINED HEREIN IS PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR OTHERWISE REPRODUCED, REPACKAGED,
FURTHER TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN WHOLE OR IN PART. IN
ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER. BY ANY PERSON WITHOUT MOODY'S PRIOR WRITTEN CONSENT. All onformaMn contaoned hereon os obtaoned by
MOODY'S from sources believed by It to be accurate and reliable, Because of the pOSSibility of human or mechanical error as well as other factors, however, such information IS proVided "as
IS" Without warranty of any kind and MOODY'S, In particular, makes no representation or warranty, express or Implied, as to the accuracy, timeliness, completeness, merchantability or fitness
for any particular purpose of any such Information, Under no circumstances shall MOODY'S have any liability to any person or entity for (a) any loss or damage In whole or In part caused by,
resulting from, or relating to, any error (negligent or otherwise) or other circumstance or contingency Within or outSide the control of MOODY'S or any of Its directors, officers, employees or
agents In connection With the procurement. collection, compilation, analYSIS, Interpretation, communication, publication or delivery of any such information, or (b) any direct. indirect.
speCial, consequential, compensatory or inCidental damages whatsoever (Including Without limitation, lost profits). even If MOODY'S IS adVised In advance of the pOSSibility of such
damages, resulting from the use of or inability to use, any such information, The credit ratings and financial reporting analYSIS observations, If any, constituting part of the information
contained herein are, and must be construed solely as, statements of opinion and not statements of fact or recommendations to purchase, sell or hold any SeCUrities, NO WARRANTY,
EXPRESS OR IMPLIED, AS TO THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR OTHER
OPINION OR INFORMATION IS GIVEN OR MADE BY MOODY'S IN ANY FORM OR MANNER WHATSOEVER. Each ratong or other opon,on must be we'ghed solely as one factor on any
Investment deCISion made by or on behalf of any user of the information contained herein, and each such user must accordingly make ItS own study and evaluation of each security and of
each Issuer and guarantor of. and each proVider of credit support for, each security that It may conSider purchaSing, holding or seiling
MOODY'S hereby discloses that most Issuers of debt seCUrities (including corporate and municipal bonds, debentures, notes and commerCial paper) and preferred stock rated by
MOODY'S have, prior to assignment of any rating, agreed to pay to MOODY'S for appraisal and rating services rendered by It fees ranging from $1,500 to approximately $2.400,000
Moody's Corporation (MCO) and ItS wholly-owned credit rating agency subs diary, Moody's Investors Service (MIS), also maintain policies and procedures to address the Independence of
MIS's ratings and rating processes, Information regarding certain affiliations that may eXist between directors of MCO and rated entities, and between entities who hold ratings from MIS and
have also publicly reported to the SEC an ownerShip Interest In MCO of more than 5%, IS posted annually on Moody's website at \MNW,moodys,com under the heading "Shareholder
Relations - Corporate Governance - Director and Shareholder Affiliation Policy"
Special Comment
October 2007
Table of Contents:
Highlights
Introduction
4
Accuracy Ratios
5
Investment-Grade Loss Rates
6
7
Average Rating Before Impairment
Rating Action Rates and Large Rating Action
Rates
9
Glossary
13
Related Research
15
Analyst Contacts:
New York
1.212.553.1653
Julia Tung
Vice President - Senior Credit Officer
The Performance of
Structured Finance Ratings:
Mid-Year 2007 Report
Highlights
This Special Comment updates Moody's structured finance rating performance
metrics as of June 2007.
As indicated in Figure 1, the one-year rating accuracy and stability of the most
recent rating cohort (ending June 2007) was lower than that of the cohort formed
six months ago, but still significantly higher compared to the long-term average
performance of all prior cohorts. The highlights of this report are:
:::
Richard Cantor
Team Managing Director
Overall, 134 structured finance securities became impaired in the first
half of 2007: 102 in US ABS, 12 in US CMBS, 14 in US RMBS, 5 in
global COOs, and one in the international structured finance sector
excluding COOs. Of these, 112 were principal impairments (suffered
principal losses or were downgraded to Ca or C), while the remaining 22
were interest impairments (experienced interest shortfalls only).
The one-year accuracy ratio for the most recent cohort was 84.6%,
which is 9.9 percentage points lower than its six months-prior level, but
1
7.5 percentage points higher than the historical average.
The five-year accuracy ratio for the most recent five-year cohort, formed
in July 2002, was 58.9%, 2.1 percentage points lower than its level of
61.0% six months ago.
:':
The one-year investment-grade loss rate was 0.05%, an increase over
the loss rate of 0.01 % for the cohort ending December 2006, but still
very low in absolute terms.
1 These performance metrics should be interpreted with caution. Some statistics are based on small samples, as the number of impairments in any
given year and any given sector is often small. Individual performance metrics may also be driven by highly correlated collateral performance
underlying multiple securitizations, particularly within certain sectors. Moreover, variations in rating performance over time may reflect either changes
in the quality of the rating process or changes in the environment that make losses more or less difficult to predict or make collateral performance more
volatile, thus resulting in either higher rating accuracy and lower rating stability at one time than another, or vice versa.
Moody's Investors Service
:,:,
The average rating during the three years prior to impairment of all securities that were impaired for the
cohort ending in June 2007 rose to Ba3, two notches above the average rating prior to impairment for the
cohort ending six months earlier, but still a notch below the historical average.
':':
The one-year rating action rate dropped to 7.4% in the most recent cohort from its six months-prior level of
7.9%, as the recent increase in the frequency of downgrades was more than offset by a decrease in the
frequency of upgrades .
.:.:
The one-year large rating action rate (rating changes of three notches or more) also declined to 3.3% from
3.5% six months ago.
':':
Performance was mixed among the various sectors of structured finance (Figure 2). The one-year
accuracy ratio for US CMBS held steady at 94%, while that of global COOs and international structured
finance excluding COOs reached all-time highs of greater than 98%. Meanwhile, the US ABS and US
RMBS one-year accuracy ratios decreased to 87% and 80%, respectively.
:':'
Structured finance rating performance metrics for the second half of 2007 will differ sharply from those in
the first half of the year in light of the large number of rating actions that have already occurred between
the end of June and early October. 2
Figure 1 - Summary of All Structured Finance Rating Performance as of June
20073
36-
Number
of New
Impairments
over
Prior 12
Months
1-Year
Accuracy
Ratio
June 2007
167
December 2006
June 2006
December 2005
Average (1993-Most Recent)
Cohort Ending Date
5-Year
Accuracy
Ratio
1-Year
InvestmentGrade
Loss Rate
MonthAverage
Rating
Before
Impairment
1-Year
Rating
Action
Rate
1-Year
Large
Rating
Action
Rate
84.6%
58.9%
0.05%
Ba3
7.4%
3.3%
107
94.5%
61.0%
0.01%
B2
7.9%
3.5%
88
95.5%
62.2%
0.00%
B1
7.0%
3.0%
84
91.6%
64.5%
0.02%
Ba2
8.4%
3.5%
88
77.1%
67.6%
0.18%
Ba2
8.1%
3.9%
In particular, between July 1, 2007 and October 11, 2007 there were 758 new downgrades to CalC (752 affecting US subprime securities - which will
impact ABS sector statistics, 1 affecting a US RMBS transaction, 4 affecting market-value COOs, and 1 affecting an aircraft lease ABS transaction).
•
These new material impairments will be included in our next structured finance default and loss study and our next performance report.
3 A glossary appears at the end of this report. The number of impairments for historical cohorts is subject to revision during each update as payment
shortfalls can be cured and past remittance or trustee reports may be revised. In addition, consistent with Moody's annual default and loss study,
Moody's now derives loss rates using loss-given-default (LGO) from principal impaired securities alone. For more details, see the entry for LGO in the
glossary. The historical average of the number of new impairments over the prior 12 months is calculated as the total number of newly impaired tranches
divided by the number of years in the sample period, and has been rounded to the nearest integer unless rounding results in zero.
2
ttr:trrOctober 2007 :':': Special Comment:':' Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
'::::::::::::::::::::::: ...
Figure 2 - Summary of Structured Finance Rating Performance by Sector as of
June 200?
Number
of New
36MonthAverage
Rating
Before
Impairment
1-Year
Rating
Action
Rate
1-Year
Large
Rating
Action
Rate
Impairments
over
Prior 12
Months
1-Year
Accuracy
Ratio
5-Year
Accuracy
Ratio
1-Year
InvestmentGrade
Loss Rate
US ABS
113
86.9%
55.7%
0.11%
Ba2
6.5%
3.6%
US CMBS
21
93.8%
86.0%
0.00%
Caa3
21.6%
8.2%
US RMBS
17
80.2%
81.6%
0.02%
Ba3
3.5%
1.6%
Global CDOs
15
98.4%
65.2%
0.00%
Caa1
7.5%
3.2%
99.8%
91.3%
0.00%
Caa3
4.0%
1.0%
Cohorts Ending June 2007
Inn SF ex. CDOs
Historical Averages Since 1993
US ABS
53
77.7%
68.5%
0.29%
Ba1
6.6%
4.0%
US CMBS
8
91.0%
87.3%
0.01%
Caa1
15.2%
5.3%
US RMBS
6
87.7%
88.2%
0.01%
Ba3
6.5%
3.1%
Global CDOs
20
76.0%
61.5%
0.38%
Ba2
10.9%
5.7%
Inn SF ex. CDOs
0.3
59.6%
75.1%
0.01%
Ba1
5.5%
1.8%
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~:~:::~~~~~~~f·october 2007
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
Introduction
In a Special Comment published in April 2003, Moody's developed a set of metrics to measure the
performance of corporate ratings with respect to the dual objectives of rating accuracy and rating stability.4
Moody's corporate rating performance report is now updated on a quarterly basis.5 Moody's first introduced
and examined its structured finance rating performance metrics in a September 2004 Special Comment
"Default & Loss Rates of Structured Finance Securities: 1993-2003," and published these performance metrics
in a stand-alone document for the first time in September 2005. The structured finance rating performance
report is now updated on a semi-annual basis.
For both the corporate and structured finance rating performance reports, the basic unit of observation is a
monthly cohort of ratings, i.e. all outstanding ratings at the beginning of a month are recorded and their
performance tracked over different time horizons. In computing rating performance metrics for structured
finance, Moody's incorporates both the default and loss severity experience of all structured finance tranches
because Moody's structured finance ratings rank order expected loss rates. In other words, Moody's
structured rating performance metrics weigh those tranches that have become materially impaired but with
lower loss severity less than those with higher loss severity.
The most important measure of rating accuracy is the accuracy ratio, which for structured finance measures
the relationship between tranche ratings and their realized loss rates.
6
This metric measures the quality of
Moody's ratings as indicators of relative expected credit loss risk.
As discussed in the April 2003 Special Comment, although relative rating accuracy is our primary objective,
Moody's recognizes that many investors are also concerned with the cardinal accuracy of the rating system.
In particular, they expect that investment-grade credits should rarely suffer credit losses and impaired
securities should normally carry low ratings well in advance of impairment. For this purpose, we regularly
track investment-grade loss rates and the average rating of securities during the 36 months prior to
impairment. Both of these measures should be low if the rating system is accurate in a cardinal sense.
We employ two measures of rating stability (or rating volatility) - the rating action rate (the frequency of rating
changes) and the large rating action rate (the frequency of rating changes of three notches or more) over a
twelve-month period. To examine how watchlist actions are used, we also report the percentages of
downgrades and upgrades preceded by watch list (review) actions in the same direction.
The data sample used in this report includes all public, 144A, and private tranches with a Moody's long-term
global debt rating in the global asset-backed securities (ABS), commercial and residential mortgage-backed
securities (CMBS and RMBS) and collateralized debt obligations (COO) sectors. Provisional ratings, credit
evaluations, and national scale ratings are not included. Pari passu tranches are collapsed into a single
tranche.
7
In addition, the following tranches are excluded: (1) tranches guaranteed by financial guarantors,
government agencies, or government-sponsored enterprises (GSEs); (2) interest-only (10) or residual
tranches; (3) repackaged securities, structured notes, structured investment vehicles, structured covered
bonds, and other credit derivative securities; and (4) deals that have all their tranche ratings linked to a single
corporate or sovereign rating.
4 See "Measuring the Performance of Corporate Bond Ratings," Moody's Special Comment, April 2003.
•
5 For the latest performance report, see "The Performance of Moody's Corporate Bond Ratings: June 2007 Quarterly Update," Moody's Special
Comment, August 2007.
•
6 The required adjustments to convert the standard default-based AR measure to a loss-based measure are discussed in "Default & Loss Rates of
Structured Finance Securities: 1993-2003," Moody's Special Comment, September 2004. The concept of accuracy ratio is also described in the glossary
at the end of this report.
•
7 More exactly, tranches carrying the same rating from the same deal, regardless of their rating levels, are collapsed into a single rating observation, with
the following exception: if two or more tranches share the same rating in the same deal, but are collateralized by distinct groups of loan pools, then the
tranches are not collapsed. Additionally, we do not review each tranche of every deal in order to determine whether it is pari passu to another tranche of
the same deal.
~~~~~~~~~:~::~~~~~~~~~·.·?ctober 2007
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
Accuracy Ratios
Figure 3 depicts the one- and five-year accuracy ratios over time with the following notable observations:
::
The one-year accuracy ratio for the overall structured finance category decreased to 84.6% from its level
six months ago of 94.5% due to an increase in the number of material impairments among securities
backed by residential mortgages within US ABS and RMBS.
:.:.
For the cohort ending June 2007, the one-year accuracy ratio for US CMBS held steady at 94%, while the
ratio for global COOs reached an all-time high of 98%.8 For US ABS and RMBS, the one-year accuracy
ratio dropped from 95% for the cohort ending January 2007 to 87% and 80%, respectively. However, the
accuracy ratio for US ABS remained above its historical average of 78% and the absolute number of new
impairments over the prior 12 months for US RMBS was still low.
:':'
The five-year accuracy ratios followed the same trend as the one-year accuracy ratios for all sectors, with
declines seen for US ABS, US RMBS, and global structured finance as a whole, while the US CMBS and
global COO accuracy ratios exhibited stability or improvement.
8 The figures for US ABS and RMBS in this study differ from those reported in studies published prior to 2007 because some of the old vintage RMBS
transactions sponsored by DLJ were reclassified into the HEL (and therefore ABS) category. Please refer to "Deal Sponsor and Credit Risk of US ABS
and MBS Securities," Moody's Special Comment, December 2006 which discusses how these transactions were backed by subprime mortgage loans by
today's standard. Please also see the glossary for the definition of RM BS, HEL, and ABS.
~~~~~~~~~~~:::~~~~~~~~~·.·?ctober 2007
:':': Special Comment ':':' Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
Figure 3 - One-Year (solid line) and Five-Year (dotted line) Accuracy Ratios
,,_,0,:
,": ,: ...,....... ,......................................"....,............,... ..
Note: At the beginning of each month, securities are grouped together by their alpha-numeric ratings to form rating cohorts. The
latest one- and five-year cohorts are formed on July 1, 2006 and July 1, 2002. Breaks in the accuracy ratio series occur when
the number of impairments for the cohort is less than or equal to one. Crosses in the one-year series and plus signs in the fiveyear series indicate that the accuracy ratio was computed from only one impaired security.
Investment-Grade Loss Rates
Figure 4 shows one-year and five-year investment-grade loss rates by sector. Note that:
~~~~~~~~~:::::~~~~~~~f·October 2007
':.:
For global structured finance, the one-year investment-grade loss rate rose to 0.05% from 0.01% six months
prior, but was still very low in absolute terms and relative to the cohorts formed between 2001 and 2004.
':':
For the cohort formed in July 2006, the one-year investment-grade loss rate was zero for US CMBS,
global COOs, and the international structured finance sector excluding COOs, and has remained zero for
at least the last 15 months for each of these three sectors.
':.:
The one-year investment-grade loss rate increased to 0.11 % from 0.03% for US ABS over a six-month
period, but was still far lower than the average over 2002-2004. The US RMBS investment-grade loss rate
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
for the July 2006 cohort was 0.03%, an increase relative to the loss rate of 0.004% six months ago, but
similar to the loss rate of 0.03% 18 months earlier.
:.:.
The five-year investment-grade loss rate for overall structured finance and all its sectors displayed a
declining trend with the exception of US RMBS. Although the five-year investment-grade loss rate
increased for US RMBS, it still remained among the lowest of all sectors.
Figure 4 - One-Year (solid line) and Five-Year (dotted line) Investment-Grade Loss Rates
...
Note: At the beginning of each month, all securities carrying an investment-grade rating are grouped together to form a
rating cohort. The latest one- and five-year cohorts are formed on July 1, 2006 and July 1, 2002.
Average Rating Before Impairment
Figure 5 presents the 36-month-average rating before impairment over time as well as the number of newly
impaired securities used to calculate the average ratings. The following observations are noteworthy:
':':
~~~~~~~~~~;~:~~~~~~~f·october 2007
The average rating before impairment for global structured finance rose to Ba3 for the most recent cohort,
two notches up from B2 six months prior, but one notch below its historical average of Ba2. However,
except for US ABS, the average ratings before impairment for all the sub-sectors of structured finance
were at or below their levels six months earlier.
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
:,:,
The 36-month average rating prior to impairment for US ABS rose two notches to Ba2 for the cohort
ending June 2007 from B1 for the cohort ending December 2006, but was still below the historical average
of Ba1.
':':
US CMBS, global COOs, and international structured finance excluding COOs all exhibited low average
ratings before impairment for the most recent cohort at Caa3, Caa1, and Caa3, respectively.
Figure 5 - 36-Month-Average Ratings before Impairment (solid line) and
Number of Newly Impaired Securities (dotted line)
:::,:
....:....... ,., ...,." ..:.L ...... , ....... , ................................. .
Note: At the beginning of each month, all securities that become impaired within the next 12-month period are grouped to
form a cohort. The latest 12-month cohort is formed on July 1, 2006. Breaks in the average rating before impairment series
occur when the number of impairments for the cohort is less than or equal to one. Crosses indicate that the average rating
was computed from only one impaired security.
~~~~~~~~~:;:::~~~~~~~f·October 2007
:':': Special Comment ':':' Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
Rating Action Rates and Large Rating Action Rates
Figure 6 reports 12-month rating action rates and large rating action rates. Figures 7 and 8 further
disaggregate rating actions into downgrades and upgrades, and Figure 9 demonstrates how frequently
downgrades and upgrades have been preceded by watchlist actions in the same direction.9 Key observations
include:
.:.:
The 12-month rating action rate in the all structured finance category was 7.4% for the July 2006 cohort,
down slightly from the rate of 7.9% six months prior and the historical level of 8.1 %.10 The large rating
action rate was also somewhat lower at 3.3%, from 3.5% six months ago and 3.9% historically.
::
The global structured finance downgrade rate increased moderately to 2.2% from 1.9% six months earlier,
but was still far below the peak of 8.7% reached by the July 2002 cohort. The large downgrade rate also
rose from 1.0% to 1.3% over the same period. The major contributor of structured finance downgrades
was US ABS, specifically securities backed by home equity loans. US ABS was the only sector which
experienced a substantial increase in the downgrade rate, including an increase in the frequency of large
downgrades.
:.:.
The overall upgrade rate for the most recent cohort was 5.2%, a drop from the upgrade rate of 6.0% six
months ago, but in line with the rate of 5.4% twelve months ago. The frequency of large upgrades also
experienced a corresponding decline to 2.0% from 2.6% six months prior and 2.1 % twelve months prior.
The US CMBS upgrade rate decreased from an all-time high of 22.5% for the January 2006 cohort to
19.5% for the July 2006 cohort. US RMBS and international structured finance excluding COOs also
experienced declines in their upgrade rates over the last six months, while the US ABS upgrade stayed
steady at 3.1% and the global COO upgrade rate ticked upwards from 4.0% to 4.3%.
::
The proportion of downgrades and upgrades that were placed on review prior to the rating action
continued a declining trend and fell to 72% for downgrades for the July 2006 cohort and 27% for upgrades.
The declines were led by a decrease in the percentage of reviewed rating changes among US ABS and
US RMBS. Meanwhile, global COOs and international structured finance excluding COOs experienced an
upward trend for reviewed downgrades with over 90% of the downgrades in the 12 months before June
2007 being placed on review for downgrade prior to the rating action.
Figure 6 -12-month Rating Action Rates (solid line) and 12-month Large
(three notches or more) Rating Action Rates (dotted line)
.. .. ,-:
,-; ,
::::,;::".".-};;
.
.
Moody's also reviewed its uses of watchlist actions on structured finance securities in a Special Comment, "Structured Finance Watchlist Resolutions:
1992-2003," June 2004.
•
10 Note that the stated rating action rate excludes rating actions affecting deals with closing dates after June 2006 because the rating must have been
outstanding as of the beginning of July 2006 in order to be counted in the calculation.
9
~~~~~~~~~:i:;:~~~~~~~~~·.·?ctober 2007
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
cO'" .,•.•••.••, ••••••••,••• :.•••••••,: •••:c.: .................................. .
":' :.. ~ ,.>-' .
Note: Rating actions include upgrades and downgrades. which are measured on the alpha-numeric (or modified) rating
scale. Rating cohorts are formed each month covering a 12-month period. The latest 12-month cohort is formed on July 1.
2006.
Figure 7 -12-month Downgrade Rates (solid line) and 12-month Large (three
notches or more) Downgrade Rates (dotted line)
,.
.',
~~~~~~~:::~~::~::~~~~f·October 2007
..
'
..
..
.'.
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
"
':'.
':'.
:.:
:.:
-;.-
';',
.-;'
.-;'
:.:
,-:'
','
,',
:
Note: Downgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month
covering a 12-month period. The latest 12-month cohort is formed on July 1, 2006.
Figure 8 - 12-month Upgrade Rates (solid line) and 12-month Large (three
notches or more) Upgrade Rates (dotted line)
Note: Upgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month
covering a 12-month period. The latest 12-month cohort is formed on July 1, 2006.
~~~~~~~:::~~~:::~~~~~f'october 2007
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
Figure 9 - Percentages of Downgrades (solid line) and Upgrades (dotted line)
Preceded by Watchlist Actions in the Same Direction
'"
.
"":;;;.
Note: At the beginning of each month, all securities that experienced downgrades (upgrades) within the next 12-month
period are grouped together to form a rating cohort. The latest 12-month cohort is formed on July 1, 2006. Gaps in the data
indicate that there were no downgrades (upgrades) during that time period.
~~~~~~~:::~!:i:~~~~~f'october 2007
::::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
Glossary
Payment Shortfall
Structured finance securities are defined as being in payment shortfall (previously called "payment default") if
they have suffered:
::
an interest shortfall, or
::
a principal write-down.
Material Impairment
Structured finance securities are defined as being in material impairment if they have:
::
sustained a payment shortfall that remained uncured, or
::
been downgraded to Ca or C.
Prepayment-related and AFC-related interest shortfalls are not considered to be material impairments, but
PIKing tranches are. Explicit principal write-downs are included whereas implicit principal write-downs or
under-collateralizations are not.
The impairment status of a security may change as it goes from cured (i.e. all outstanding shortfalls and losses
were repaid in full) to uncured (i.e. positive interest shortfalls or principal losses outstanding), or vice versa. If
any securities rated Ca or C but not in payment shortfall are upgraded, they are considered to be no longer in
material impairment. Securities rated Ca or C that were not upgraded are in material impairment even if their
payment shortfalls have been cured. Finally, securities with very minor shortfalls or losses are excluded.
A security is called "newly impaired" in a given period if it had no outstanding payment shortfalls and was not
downgraded to Ca or C at the end of the prior period, but experienced at least one of these two credit events
at the end of the given period.
Principal Impairment
This refers to materially impaired securities that have suffered principal write-downs or principal losses, or
have been downgraded to Ca or C (with or without experiencing principal write-downs or losses). In particular,
if a security had experienced principal write-down/loss or was downgraded to Ca or C, it is called a principal
impairment regardless of whether it had experienced interest shortfalls.
Interest Impairment
This refers to materially impaired securities that have experienced only interest shortfalls, no principal losses,
and were not downgraded to Ca or C.
Investment-Grade (IG) and Below Investment-Grade (BIG)/SpeculativeGrade (SG) Ratings
Investment-grade ratings refer to Aaa, Aa1, Aa2, Aa3, A1, A2, A3, Baa1, Baa2, and Baa3. Below investmentgrade or speculative-grade ratings refer to Ba1, Ba2, Ba3, B1, B2, B3, Caa1, Caa2, Caa3, Ca, and C.
Loss-Given-Default (LGD) or Loss Severity Rate
The loss-given-default, also known as LGD or loss severity rate, of an impaired security is measured by the
sum of the present values of net losses, including both interest shortfalls and principal losses, discounted by
the security's coupon rate and expressed as a percentage of the principal balance on a certain reference date.
There are typically three types of reference dates used in the calculation of LGD: origination, the time of
impairment, and a cohort formation date.
Final LGD on impaired securities with a zero outstanding principal balance are typically known. For impaired
securities with positive principal balances or impaired securities with incomplete loss data, their LGD need to
~~~~~~~:::~~::~::~~~~~~·.·?ctober 2007
::::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
11
be estimated.
In addition, consistent with Moody's annual default and loss study, Moody's now derives loss
rates using LGD from principal impaired securities alone. In almost all cases, securities with only interest
shortfalls (interest impairments) end up being cured or suffering principal writedowns or being downgraded to
CalC (principal impairments). Interest impairments have frequently been cured, whereas cures on principal
impairments have been rare. As a result, we do not attempt to project final LGD on interest impaired securities
until they either become cured, in which case they drop out of the LGD data sample with an LGD of zero, or
become principal impaired, in which case a projection of the final LGD is made if the impairment is not
resolved at that time.
Accuracy Ratio (AR)
An accuracy ratio based on the loss experience of structured finance securities (or an accuracy ratio adjusted
for loss-given-default, or LGD) is the ratio of the area between a loss-based cumulative accuracy profile (CAP)
curve and the 45-degree line to the maximum possible area above the 45-degree line. The loss-based CAP
curve (or a CAP curve adjusted for LGD) plots, for each rating category, the proportion of the losses of all
impaired securities accounted for by securities with the same or lower rating against the proportion of all
securities in the sample population with the same or lower rating. 12
To calculate accuracy ratios, rating cohorts are formed for each calendar month over the study period so that
all outstanding securities in the sample population are grouped together by their alpha-numeric ratings at the
beginning of the month. For each monthly rating cohort, we determine the number of securities that became
impaired within one or five years of the cohort formation date and their loss severity rates as a percentage of
the cohort date balance. Cumulative shares of securities rank-ordered by rating are calculated for the
universe of all securities and the universe of impaired securities, respectively, and based on this, a CAP curve
is plotted. Note that only LGD for principal impaired securities are used in the calculation. Please see the
definition of LGD for further details.
The CAP curve adjusted for LGD is also known as a "power curve" because it shows how effective a rating
system is at differentiating between securities that have sustained high losses from securities that have
sustained low or no losses. The metric is defined relative to the distribution of ratings in the population. The
accuracy ratio is an effective way to summarize the CAP curve into a single number. The accuracy ratio is
zero if the CAP curve collapses to the 45-degree line, suggesting that all impaired securities are randomly
distributed throughout the population without regard to rating.
Because we use only data observations that span the time horizon under consideration, the latest cohort for a
one-year horizon is formed on July 1, 2006, while for a five-year horizon the latest cohort is formed on July 1,
2002.
Investment-Grade Loss Rate
The one-year investment-grade loss rate is calculated as follows. First, for a given cohort, we calculate the
LGD as a share of the tranche balance at the cohort date for each security that carried an investment-grade
rating at the cohort formation date and became impaired within a 12-month period after the cohort date. We
then take the sum of these LGD rates and divide by the total number of investment-grade securities
outstanding as of the cohort formation date. Note that the LGD rate is not weighted by the total dollar volume
of outstanding securities in the cohort. Also note that only LGD for principal impaired securities are used in the
calculation. Please see the definition of LGD for further details. The five-year investment-grade loss rate is
calculated similarly.
Average Rating before Impairment
The rating of an impaired security is measured every month for 36 months prior to impairment. These 36 rating
measurements are averaged together to create one representative number for each impaired security. For a
11 See "Default & Loss Rates of Structured Finance Securities: 1993-2006," Moody's Special Comment, April 2007 and "Measuring Loss-Given-Default
for Structured Finance Securities: An Update," Moody's Special Comment, December 2006 for discussions of the characteristics and estimation methods
clLGD.
.
12 For an illustration of the CAP curve adjusted for LGD, see "Default & Loss Rates of Structured Finance Securities: 1993-2003," Moody's Speciaf
Comment, September 2004.
~~~~~~~:::~~:::::~~~~~~·.·?ctober 2007
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
particular cohort, the average rating before impairment is the weighted average of these average ratings for
each security that became impaired within 12 months after the cohort formation date, where the weight for
each security is the LGD rate of the tranche as a share of its original balance. This weighting scheme will
place greater emphasis on the average ratings of impaired tranches with higher LGD over impaired tranches
with lower LGD. 13 Note that only LGD for principal impaired securities are used in the calculation. Please see
the definition of LGD for further details.
Rating Action Rate
The rating action rate is defined as the number of securities that experienced a rating change within a year
after cohort formation divided by the total number of securities outstanding at the cohort formation date (the
beginning of each month). Rating changes are measured on the alpha-numeric (or modified) rating scale and
are based on comparing the rating at the beginning and end of the time period under consideration. However,
if a rating was withdrawn by the end of the time period, then the rating prior to withdrawal is used as the end
rating. Note that a security will only be counted if it was outstanding as of the cohort formation date.
Downgrade rates and upgrade rates are measured similarly based on downgrade rating actions and upgrade
rating actions, respectively.
Large Rating Action Rate
A large ration action is said to occur if a rating action (or cumulative rating actions) cause(s) a security's rating
to change by three or more notches within a year after cohort formation. The large rating action rate is the
number of such securities divided by the total number of securities outstanding at the cohort formation date.
Large downgrade rates and large upgrade rates are measured similarly based on large downgrade rating
actions and large upgrade rating actions, respectively.
Percentage of Downgrades (Upgrades) Preceded by Watchlist Actions in the
Same Direction
This metric is defined as the total number of downgraded (upgraded) securities that were placed on the
watch list in the same direction before they were downgraded (upgraded), divided by the total number of
securities that were downgraded (upgraded) within 12 months after the cohort formation date.
ABS
ABS stand for asset-backed securities. This structured finance sector includes securities backed by home
equity loans (HEL) and both traditional asset types such as auto loans, credit card receivables, student loans,
and manufactured housing loans, and non-traditional asset types such as mutual fund fees, tax liens, tobacco
settlement payments, and intellectual property.
HEL
The home equity loan or HEL sector includes securities backed by subprime (B&C) mortgage loans, home
improvement loans, high loan-to-value (high LTV) loans, home equity lines of credit (HELOCs), and closedend second-lien loans, as well as net interest margin (N 1M) securitizations. It does not include securities
backed by Alt-A mortgages, which are included in the RMBS sector. HEL is part of the ABS sector.
Prior to 1998, RMBS collateral was generally defined as first-lien residential mortgages, regardless of the
credit quality of the borrower. HEL collateral generally included junior liens such as HELOCs or closed-end
seconds. However, as subprime lending became more prevalent, the market shifted its definition such that
HEL encompassed subprime first-lien residential mortgages while RMBS included first-lien mortgages made to
higher quality borrowers. Since 1998 and especially in the last five years, a deal classified as RMBS by
Moody's is generally backed by prime or Alt-A quality first-lien residential mortgages, while a deal classified as
13 We began using LGD rates as weights in computing an average rating before impairment in the full-year 2005 structured finance performance report.
Ideally, LGD rates should be calculated as a percentage of the principal balance outstanding for each month in the 36 months prior to the impairment,
and ratings should then be weighted by these monthly LGD rates. Practically, however, it does not make any material difference in the average rating
number whether we use LGD rates as a share of impairment-date balance, original balance, or monthly principal outstanding. Since the monthly LGD
rates are very time consuming to compute due to amortization, we use the LGD rate as a share of original balance as the weight variable.
~~~~~~~:::~~~:~::~~~~~~·.·?ctober 2007
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
HEL can be backed by subprime first-lien mortgages or junior liens. Therefore, a subprime deal which would
be classified as HEL today may have been classified as RMBS in the past.
CDOs
COOs stand for collateralized debt obligations. Derivative securities such as structured notes and repackaged
securities are not considered to be part of this sector. Commercial real estate (CRE) COOs, where 70% or
more of the collateral is comprised of CRE loans, are classified as CMBS. If the collateral backing the
transaction contains less than 70% CRE loans, then the deal is classified as a COO.
CMBS
CMBS stand for commercial mortgage-backed securities. Commercial real estate (CRE) COOs, where 70% or
more of the collateral is comprised of CRE loans, are classified as CMBS. If the collateral backing the
transaction contains less than 70% CRE loans, then the deal is classified as a COO.
RMBS
RMBS stand for residential mortgage-backed securities. The large majority of these securities are backed by
first-lien prime mortgages, but some are backed by Alt-A mortgages. For further details, see the definition of
HEL.
All Structured Finance Securities
All structured finance captures global structured finance securities in four major sectors: ABS, COO, CMBS,
and RMBS.
U.S. Structured Finance
u.s. structured finance securities are denominated in U.S. dollars and issued in the U.S. market or
denominated in Canadian dollars and issued in Canada. In cases where the source of the underlying
collateral and the denomination of the securities crossed multiple countries/regions, deals are classified by the
location at which they are monitored.
International Structured Finance excluding CDOs
This refers to non-COO structured finance securities that are not denominated in U.S. dollars and issued in the
U.S. market and not denominated in Canadian dollars and issued in Canada. The majority of the securities in
this sector are issued in Europe, the Middle East, and Africa (EMEA); the rest are issued in the Asia Pacific
region and Latin America.
~~~~~~~:::~~::~:;~~~~~~·.·?ctober 2007
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
Related Research
':':
The Performance of Structured Finance Ratings: Full-Year 2006 Report, May 2007 (103017)
::
The Performance of Structured Finance Ratings: Mid-Year 2006 Report, September 2006 (99034)
':::
The Performance of Structured Finance Ratings: Full-Year 2005 Report, May 2006 (97346)
::
The Performance of Structured Finance Ratings: Mid-Year 2005 Report, September 2005 (94463)
':::
Default & Loss Rates of Structured Finance Securities: 1993-2006, April 2007 (102733)
::
Default & Loss Rates of Structured Finance Securities: 1993-2005, April 2006 (97234)
':::
Default & Loss Rates of Structured Finance Securities: 1993-2004, July 2005 (93653)
::
Default & Loss Rates of U.S. COOs: 1993-2003, March 2005 (91692)
::
Default & Loss Rates of Structured Finance Securities: 2004 First Half Update, January 2005 (90843)
::
Default & Loss Rates of Structured Finance Securities: 1993-2003, September 2004 (88692)
::
Measuring Loss-Given-Default for Structured Finance Securities: An Update, December 2006 (101284)
::
Measuring Loss Severity Rates of Defaulted Residential Mortgage Backed Securities: A Methodology,
April 2004 (86769)
.:.:
Payment Defaults and Material Impairments of U.S. Structured Finance Securities: 1993-2002, December
2003 (80247)
':::
Structured Finance Rating Transitions: 1983-2006, January 2007 (101833)
::
Structured Finance Rating Transitions: 1983-2006 H1, August 2006 (98577)
':::
Structured Finance Rating Transitions: 1983-2005, February 2006 (96533)
::
Structured Finance Rating Transitions: 1983-2004, February 2005 (91392)
':::
Structured Finance Rating Transitions: 1983-2003, February 2004 (81239)
::
Structured Finance Rating Transitions: 1983-2002, January 2003 (77291)
::
European Structured Finance Rating Transitions: 1988-2006, February 2007 (102033)
::
European Structured Finance Rating Transitions: 1988-2005, February 2006 (96706)
::
Japanese Structured Finance Rating Transitions: 1994-2006, March 2007 (102509)
::
Asia Pacific (ex-Japan) Structured Finance Rating Transitions: 1990-2006 (102095)
::
Deal Sponsor and Credit Risk of U.S. ABS and MBS Securities, December 2006 (100872)
::
The Relationship between Par Coupon Spreads and Credit Ratings in US Structured Finance, December
2005 (95494)
':':
Structured Finance Watchlist Resolutions: 1992-2003, June 2004 (87305)
:':'
The Performance of Moody's Corporate Bond Ratings: June 2007 Quarterly Update, August 2007
(104308)
::
Measuring the Performance of Corporate Bond Ratings, April 2003 (77916)
':::
Guide to Moody's Default Research: April 2007 Update, April 2007 (102661)
To access any of these reports, click on the entry above. Note that these references are current as of the date of publication
of this report and that more recent reports may be available. All research may not be available to all clients.
~~~~~~~:::~~~:l~~~~~~~·.·?ctober 2007
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
To order reprints of this report (100 copies minimum), please caI/1.212.553.1658.
Report Number: 105390
Julia Tung
Fabian Alvarez
© Copyright 2007, Moody's Investors Service, Inc. and/or its licensors and affiliates including Moody's Assurance Company, Inc. (together, "MOODY'S"). All rights
reserved. ALL INFORMATION CONTAINED HEREIN IS PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR
OTHERWISE REPRODUCED, REPACKAGED, FURTHER TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED
FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN WHOLE OR IN PART, IN ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER, BY ANY
PERSON WITHOUT MOODY'S PRIOR WRITTEN CONSENT. All information contained herein is obtained by MOODY'S from sources believed by it to be accurate
and reliable. Because of the possibility of human or mechanical error as well as other factors, however, such information is provided "as is" without warranty of any
kind and MOODY'S, in particular, makes no representation or warranty, express or implied, as to the accuracy, timeliness, completeness, merchantability or fitness
for any particular purpose of any such information. Under no circumstances shall MOODY'S have any liability to any person or entity for (a) any loss or damage in
whole or in part caused by, resulting from, or relating to, any error (negligent or otherwise) or other circumstance or contingency within or outside the control of
MOODY'S or any of its directors, officers, employees or agents in connection with the procurement, collection, compilation, analysis, interpretation, communication,
publication or delivery of any such information, or (b) any direct, indirect, special, consequential, compensatory or incidental damages whatsoever (including without
limitation, lost profits), even if MOODY'S is advised in advance of the possibility of such damages, resulting from the use of or inability to use, any such information.
The credit ratings and financial reporting analysis observations, if any, constituting part of the information contained herein are, and must be construed solely as,
statements of opinion and not statements of fact or recommendations to purchase, sell or hold any securities. NO WARRANTY, EXPRESS OR IMPLIED, AS TO
THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR
OTHER OPINION OR INFORMATION IS GIVEN OR MADE BY MOODY'S IN ANY FORM OR MANNER WHATSOEVER. Each rating or other opinion must be
weighed solely as one factor in any investment decision made by or on behalf of any user of the information contained herein, and each such user must accordingly
make its own study and evaluation of each security and of each issuer and guarantor of, and each provider of credit support for, each security that it may consider
purchasing, holding or selling.
MOODY'S hereby discloses that most issuers of debt securities (including corporate and municipal bonds, debentures, notes and commercial paper) and
preferred stock rated by MOODY'S have, prior to assignment of any rating, agreed to pay to MOODY'S for appraisal and rating services rendered by it fees ranging
from $1,500 to approximately $2,400,000. Moody's Corporation (MCa) and its wholly-owned credit rating agency subsidiary, Moody's Investors Service (MIS), also
maintain policies and procedures to address the independence of MIS's ratings and rating processes. Information regarding certain affiliations that may exist
between directors of MCa and rated entities, and between entities who hold ratings from MIS and have also publicly reported to the SEC an ownership interest in
MCa of more than 5%, is posted annually on Moody's website at www.moodys.com under the heading "Shareholder Relations - Corporate Governance - Director
and Shareholder Affiliation Policy."
Moody's Investors Service
~~~~~~~:::~!;~::~~~~f'october 2007
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2007 Report
Special Comment
November 2008
Table of Contents:
Highlights
Introduction
4
Accuracy Ratios
5
Investment-Grade Loss Rates
7
Average Rating Before Impairment
9
Rating Action Rates and Large Rating Action
Rates
11
Appendix: Description of Data Sample and
Glossary
20
Moody's Related Research
25
The Performance of
Structured Finance Ratings:
Mid-Year 2008 Report
Highlights
This Special Comment updates Moody's structured finance rating performance
metrics as of June 2008. The highlights of this report are:
:':
Overall, 4,551 structured finance securities became impaired in the first
half of 2008: one in US ABS excluding HEL, 31 in US CMBS, 1,302 in
US HEL, 1,348 in US RMBS, 1,854 in global COOs, three in the
international structured finance sector excluding COOs and SIVs and
Other SF, and 12 in SIVs and Other SF. Of these, 4,393 were principal
impairments (suffered principal losses or were downgraded to Ca or C),
while the remaining 158 were interest impairments (experienced interest
shortfalls only).
:':
For global structured finance, the one-year accuracy ratio decreased to
59.7% from its level six months ago of 75.6% and its level twelve
months ago of 90.7% (Figure 1).1 However, excluding the most
troubled sectors - structured finance COOs (SF COOs), the SIVs and
Other SF category, and US HEL and RMBS securitized between 2005
and 2007 - the one-year accuracy ratio was higher at 83.2%.
:::
The five-year accuracy ratio, which measures the performance of
ratings that were outstanding five years ago and hence does not
incorporate the performance of the more recent vintages and more
poorly performing mortgage-related securities, was 78.1 %, 2.3
percentage points lower than its level of 80.4% six months ago.
:::
The one-year investment-grade loss rate jumped to 4.05% for the cohort
ending June 2008. This represents a new historical high and a four-fold
increase from its six months-prior rate.
Analyst Contacts:
New York
1.212.553.1653
Julia Tung
Vice President - Senior Credit Officer
Nicolas Weill
Structured Finance Chief Credit Officer
1 These performance metrics should be interpreted with caution. Some statistics are based on small samples, as the number of impairments in any
given year and any given sector is often small. Individual performance metrics may also be driven by highly correlated collateral performance
underlying multiple securitizations, particularly within certain sectors. Moreover, variations in rating performance over time may reflect either changes
in the quality of the rating process or changes in the environment that make losses more or less difficult to predict or make collateral performance more
volatile, thus resulting in either higher rating accuracy and lower rating stability at one time than another, or vice versa.
Moody's Investors Service
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
:,:,
The average rating during the three years prior to impairment of all impairments occurring during the 12month period ending June 2008 rose to Baa2, one notch above the average rating prior to impairment for
the cohort ending six months earlier, and four notches above the level for the cohort ending twelve months
earlier.
,:,:
The one-year rating action rate also reached a new record of 23.7%, almost four times higher than the
historical average of 6.1 %. Meanwhile, the frequency of large rating actions (rating changes of three
notches or more) was 20.6%, only slightly less than the overall rating action rate. Both increases were
caused by a surge in the downgrade rate, particularly for US HEL, US RMBS, and global COOs.
::':
US HEL, US RMBS, and global COOs all experienced declines in their one-year accuracy ratios,
increases in their one-year investment-grade loss rates, and higher than average ratings before
impairment (Figure 2). In contrast, US ABS excluding HEL, US CMBS, and the international structured
finance sector excluding COOs and Other SF, all displayed one-year accuracy ratios in excess of 97% and
investment-grade loss rates of zero.
Figure 1 - Summary of Global Structured Finance Rating Performance as of June 2008 2
36-
Cohort Ending Date
Number of
New
Impairments over
Prior 12
Months
1-Year
Accuracy
Ratio
5-Year
Accuracy
Ratio
1-Year
InvestmentGrade
Loss Rate
MonthAverage
Rating
Before
Impairment
1-Year
Rating
Action
Rate
1-Year
Large
Rating
Action
Rate
Global Structured Finance
June 2008
6538
59.7%
78.1%
4.05%
Baa2
23.7%
20.6%
December 2007
2135
75.6%
80.4%
0.98%
Baa3
9.6%
6.4%
195
90.7%
77.4%
0.04%
Ba3
4.6%
2.1%
June 2007
December 2006
107
96.4%
78.7%
0.01%
B2
4.8%
2.1%
Average (1993-Most Recent)
510
76.1%
83.8%
0.42%
Baa2
6.1%
3.5%
10.8%
Global Structured Finance excl SF CDOs, SIV and Other SF, and '05-'07 vintage US HEL 8: RMBS
June 2008
416
83.2%
82.1%
0.22%
Ba1
13.9%
December 2007
221
June 2007
125
87.2%
82.1%
0.11%
Ba2
6.6%
3.4%
90.7%
78.2%
0.05%
B1
6.4%
2.7%
December 2006
99
95.6%
79.5%
0.01%
B2
6.2%
2.7%
Average (1993-Most Recent)
106
87.5%
84.4%
0.10%
Ba2
5.3%
2.6%
2 A glossary appears at the end of this report. The number of impairments for historical cohorts is subject to revision during each update as paymen .
shortfalls can be cured and past remittance or trustee reports may be revised. In addition, consistent with Moody's annual default and loss study,:
Moody's now derives loss rates using loss-given-default (LGO) from principal impaired securities alone. The historical average of the number of new
impairments over the prior 12 months is calculated as the total number of newly impaired tranches divided by the number of years in the sample period,:
and has been rounded to the nearest integer unless rounding results in zero.
.
~~~~~~~~~:::;:~~~~~~~~~·.·.November 2008
:':': Special Comment ':':' Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Introduction
In a Special Comment published in April 2003, Moody's developed a set of metrics to measure the
performance of corporate ratings with respect to the dual objectives of rating accuracy and rating stability.3
Moody's corporate rating performance report is now updated on a quarterly basis.4 Moody's first introduced
and examined its structured finance rating performance metrics in a September 2004 Special Comment
"Default & Loss Rates of Structured Finance Securities: 1993-2003," and published these performance metrics
in a stand-alone document for the first time in September 2005. The structured finance rating performance
report is now updated on a semi-annual basis.
For both the corporate and structured finance rating performance reports, the basic unit of observation is a
monthly cohort of ratings, i.e. all outstanding ratings at the beginning of a month are recorded and their
performance tracked over different time horizons. In computing rating performance metrics for structured
finance, Moody's incorporates both the default and loss severity experience of all structured finance tranches
because Moody's structured finance ratings rank order expected loss rates. In other words, Moody's
structured rating performance metrics weigh those tranches that have become materially impaired but with
lower loss severity less than those with higher loss severity.
The most important measure of rating accuracy is the accuracy ratio, which for structured finance measures
5
the relationship between tranche ratings and their realized loss rates. This metric measures the quality of
Moody's ratings as indicators of relative expected credit loss risk.
As discussed in the April 2003 Special Comment, although relative rating accuracy is our primary objective,
Moody's recognizes that many investors are also concerned with the cardinal accuracy of the rating system.
In particular, they expect that investment-grade credits should rarely suffer credit losses and impaired
securities should normally carry low ratings well in advance of impairment. For this purpose, we regularly
track investment-grade loss rates and the average rating of securities during the 36 months prior to
impairment. Both of these measures should be low if the rating system is accurate in a cardinal sense.
We employ two measures of rating stability (or rating volatility) - the rating action rate (the frequency of rating
changes) and the large rating action rate (the frequency of rating changes of three notches or more) over a
twelve-month period. To examine how watchlist actions are used, we also report the percentages of
downgrades and upgrades preceded by watch list (review) actions in the same direction.
Note that the criteria used to create the data set for this report have changed from prior performance studies.
The most notable changes are that pari passu tranches are no longer collapsed and wrapped tranches are
included. For a more detailed description of the data sample, please see the Appendix.
See "Measuring the Performance of Corporate Bond Ratings," Moody's Special Comment, April 2003.
•
4 For the latest performance report, see "The Performance of Moody's Corporate Debt Ratings: September 2008 Quarterly Update," Moody's Special
Comment, October 2008.
•
5 The required adjustments to convert the standard default-based AR measure to a loss-based measure are discussed in :~Q§f§!J.I.L§.<...~Q.~.§.B§l§§.. Q.f
Structured Finance Securities: 1993-2003," Moody's Special Comment, September 2004. The concept of the accuracy ratio is also described in the
glossary at the end of this report.
.
3
~~~~~~~~~:~::~~~~~~~~~·.·.November 2008
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Accuracy Ratios
Figure 3 depicts the one- and five-year accuracy ratios over time with the following notable observations:
::
For global structured finance, the one-year accuracy ratio decreased to 59.7% from its level six months
ago of 75.6% and its level twelve months ago of 90.7%. The decline was caused by a massive increase in
the number of material impairments among US HEL and US RMBS securities issued between 2005 and
2007, and among SF COOs over the last year. Excluding these securities as well as the Other SF sector,
the one-year accuracy ratio for the cohort ending June 2008 was 83.2%.
::
US HEL and global COOs both experienced steep declines in their one-year accuracy ratios compared to
their levels six months prior and historically. The one-year accuracy ratio for US RMBS also fell
approximately five percentage points from 92.3% for the cohort ending December 2006 to 87.4% for the
cohort ending June 2008.
::
In contrast, the one-year accuracy ratios for US ABS excluding HEL, US CMBS, and international
structured finance excluding COOs and Other SF were all very high at above 97% for the cohorts formed
in July 2007. Excluding deals that closed between 2005 and 2007, the one-year accuracy ratio for US
RMBS was also close to 97%.
::
Unlike their one-year counterparts, there were no dramatic movements in the five-year accuracy ratios
between the January 2003 cohort and the July 2003 cohort for global structured finance or any of its subsectors besides global COOs. This statistic demonstrates performance on a lagged basis and the effects
of the recent dramatic increase in material impairments have not been incorporated yet.
Figure 3 - One-Year (yel/ow line) and Five-Year (blue line) Accuracy Ratios
1~::;~: ~.:~,t~~L~~~:::::~::;.:2:~::·a". &mj~~·~/
i(:%
+-----------------':.
/.,;% + - - - - - - - - - - - - ~~-,~~~-:
N::%
3,;'~';
,',;'~';
'(;~".
::J::%
+-------------+-------------+-------------+-------------+-------------::~.:
:::~:
.(~:
.(~:
8"~
}~
.~(p.:.
-))';,s
A)%
·1:f".
+-------------+-------------+-------------+-------------+-------------+--------------
'.'
~::
$":
(~::
~~
t~
::~,
:~;
:~;
.';:
.';:
.:~~
.:~~
'0;'
+------------}~
'''{i':<.
:~;
~-~--c:'c--~----~~~
,.co:
7U",. + - - - - - - - - - - - - - ':l-?:~~
'.'
;:::';'~'; +--~':,::i_"t.::~.:~::::::::········.--·.:o.·.........;:.'-. . ~·
i'(;~". +------··--"··,···::: ...... ::'.:"'1."'·::·::"-.·/-·:-·-:-- ;-~;'::::' + - - - - - - - - - - - - ,":.;';'. + - - - - - - - - - - - - ~~~:.:.~. + - - - - - - - - - - - - "K:% + - - - - - - - - - - - - :::,;% + - - - - - - - - - - - - ::.::%
... ~
·.·.·...::.:;:,,.::6:.=--/_ __
,~)%
C:
}~
t:~
}~
t~
:~~
,.C:
}~
1(iP,-;.
;.· ·. . .":. .
(:
,,)
~~---~=~ ~---~-.
~:: s~~.···
:.::.t,:,:L~~
";0% I-.c~."i:-- - - - - - - - - - - ~~f:;. :~c-:-+::-- - - - - - - - - - - ,~~%
3-:f;-:.
::-0%
~ty:_::,
c-;':""
. ...:~::-----------.'
":::;~.::
+-------------+-------------~:~
c~
"
._;.
r-)
._;.
"
r-)
,-~.
~~
Note: At the beginning of each month, securities are grouped together by their alpha-numeric ratings to form rating cohorts.
The latest one- and five-year cohorts are formed on July 1, 2007 and July 1, 2003. Breaks in the accuracy ratio series occur
when the number of impairments for the cohort is less than or equal to one. Crosses in the one-year series and plus signs in
the five-year series indicate that the accuracy ratio was computed from only one impaired security.
~~~~~~~~~~~:::~~~~~~~~~·.·.November 2008
::::: Special Comment ':::' Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Figure 3 (continued) - One-Year (yel/ow line) and Five-Year (blue line) Accuracy Ratios
j~;,~~ f<:;;:.:::-,-:-.,,::::.~:::7.:::.«,~/.:.~::::~.::"":";,;,:. ,.::.::=::::
:~·::::v·:::c:::··:/::.;·2
··:C·o,'c-;.- ~:,"'.
i:":":;:
+----------------'
;x)'",:, + - - - - - - - - - - - - ~,)S'. + - - - - - - - - - - - - -
+ - - - - - - - - - ...,.,.--'- ...::-'/-+-------------
/0'%,
/::f~t
3D',:,· + - - - - - - - - - - - - -
;}:.:".,. + - - - - - - - - - - - - ,:,;"0 + - - - - - - - - - - - - :;<;% + - - - - - - - - - - - - :.":::".,. + - - - - - - - - - - - - ',,":;: + - - - - - - - - - - - - -
:'('%
i~j':_:)
+------------+------------,"~~
'~:
"~~
e
.;~~
:..:'~
:~~
:.
:'~;
."~~~
i:;,
~::,:
~:~~
.~
.~
i~:
~~~
.~~
.C
~ tXY:{ -----:;-;'.~::.:.:.:.::::..:._....:.:.~ ::: .....;:.......::_:.:..:.:.:.:.::..:.... ,...
',i/'·,. - ' , '
3D',:,· + - - - - - - - - - - - - -
+------------;~rr:_:) + - - - - - - - - - - - - ,'-i}':", + - - - - - - - - - - - - ··~O%, + - - - - - - - - - - - - :n',:, + - - - - - - - - - - - - :,(.:S'. + - - - - - - - - - - - - iO',':' + - - - - - - - - - - - - 10%
~~':
~':
'.
G, ,
.';:
.';:
.';:
~.:
;;:
..
, ..
;;:
:~:
+-----------::----'
~xr:{· +---4\:~:.;.~:..~. i:-..... _ _::.~:~:~~
/D':-:) +------"'~.~:·~---'\········,~.;~·:;-;::~2~-:-~:-
,':;,)%
.....;:................. .~p
...:~.-"'.':..:-.- - - : c -
~~:r% +-----------':---~c_
·:n·,:, + - - - - - - - - - - - - - :
J,s'. + - - - - - - - - - - - - - - - '
?()",:,
+-----.:::~:~-----------'
1('S'.
+------------~:~
,-r-:
'0;'
'~~:':S'.
-,---------;-;
Si
..:~~
~
,::if:",
?:
?:
?:
.~;
U
S
:3~
:3~
,.
:..:)
:3~
',:::".,. + - - - - - - - - - - - .::;::.::;~~ + - - - - - - - - - - - - - - - '
:K;~;. + - - - - - - - - - - - - ?;::-:,~ + - - - - :::.:".,. + - - - - - - - - - - - - -
.~~
.
(~
~=;:.::;~~ +------:'----'-----"------~-
~
+ - - - - - - -...--:
.. ~::~{--. .- - :•...::::
~'.)
~~;~ t====~_·:::_L:'_':,···''''··\;=J:::::fLJ-
..:~~
~tXy;{
h~
g,
.C
.C
.~~
.~~
...
~;,
c~
.--:.
;~~
~':
\~~
._)
:0
"'I
~':
\~~
._)
:-.
"'I
~':
\~~
._)
-,---------------
',i/'·,. + - - - - - - - - - - - - ,;<;%
+------------c----
F:·'·,.
+------------':---~
+------~c''--------
+------------..,'if:'·, + - - - - - - - - - - - - ·~;O%, + - - - - - - - - - - - - ?()",:, + - - - - - - - - - - - - iD% + - - - - - - - - - - - - :~rr:_:)
..
i~:
~~~~~~~~~:::::~~~~~~~f·.November 2008
(~::
.C
tU~~\ +-_ _ _ _ _ _L.:~
/D':-:)
c.'
tq
==.,-------'c·.-·:··'O-o~:_:_:······
:~.
~~::
~tXy:{ -,-----------~_____c~c_
:..:'~
~-:~-n:) +--------'.::"--:~~::-:'-~:~...-..-..-....-.--.-.-.,~:-~-----,,:,---:
..._.......:{:o--.._.. - - -
;'"
':':.':" + - - - - - - - - - - - - i(;% + - - - - - - - - - - - - :::;::".,. + - - - - - - - - - - - - ,-:z;~;. + - - - - - - - - - - - - <:",~ + - - - - - - - - - - - - ::":"": + - - - - - - - - - - - - ::,</'.,. + - - - - - - - - - - - - ·,,;"0 + - - - - - - - - - - - - -
'
'~~)S'. -,------------~=c_..... ,..
,..... ",
.......~:~
\:,'
,'.
~~~
,'\',
(.~
.C
.~~
~~:
t:~~
:.:~
.C
~':
.~~
\~)
._)
',:::".,. + - - - - - - - - - - - - .::;::.::;~~ + - - - - - - - - - - - - :::::-:,~ + - - - - - - - - - - - - ',;::-:,~
+------------2~:
((~:
c(
.:~:
.:~:
.:~:
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
:~.
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Investment-Grade Loss Rates
Figure 4 shows one-year and five-year investment-grade loss rates by sector. Note that:
::
For global structured finance, the one-year investment-grade loss rate rose precipitously to 4.0% for the
cohort ending June 2008, four times higher than the loss rate of 1.0% six months prior and roughly 10 times
higher than the historical average of 0.4%. However, excluding SF COOs, the Other SF category, and US
HEL and US RMBS from the 2005 to 2007 vintages, the one-year investment-grade loss rate was only 0.2%.
::
The one-year investment-grade loss rate also grew to record-breaking highs for US HEL, US RMBS,
global COOs, and Other SF. Global COOs experienced the most striking increase, but most of this can be
attributed to the poor performance of SF COOs. Excluding this COO deal type, the one-year COO
investment-grade loss rate was 0.3%.
::
For the cohort formed in July 2007, the one-year investment-grade loss rate was zero for US ABS
excluding HEL, US CMBS, and the international structured finance sector excluding COOs and Other SF.
::
Because the five-year investment-grade loss rate is a lagging indicator, it still did not show the dramatic
increase displayed by the one-year loss rate.
Figure 4 - One-Year (yellow line) and Five-Year (blue line) Investment-Grade Loss Rates
G::)~.;,;.~~ :S~.:'l.".:~,'Jr:=;:-,j :::~):;.\n'.;;;=. 8;:;' ~:F c::x:~. O~.:~~"'::· ~:F.
C:.. :;% , - - - - - - - - - - - - - - -
'F::%
+--------------:
3,;:~';
+---------------;:.
..
:~ ~;.::::.
~ -?,::~.
:;!~::'3 'O~~~'05-' \·'l)·'~.:;~g::-'!
u=; :-~~. -& ~-:::~~::~~~,
+--------------,-
~ ~:y +-_ _ _ _ _~:,.".=·'······o ...= - - - - ' - '::.:;% -k:.:.:;::.;:-:.;.:.:::::;*::::. ::;.
..../'+"'•••..C'-P
••:'••.•~
.••.i... ::.:;;:::
............~:
........~......
;....
-
~~:
:-.:":
(.~
,C
~~:
~~i
;~~
~':
\~~
')
r
~'~'.~
~~:
,~;
::.:)%
~:...:;i·:·:\.L/-;·_ _---;:c··""···:"....:·:•
;~:.~;.::::'
,~." ········:~~-;A~"-··~~~~~;.o'
.... ~. ,..~
.0;
.•.•..- .•
~:~
~:;;
;:'~.'
:~~::
:'3
"~~
"~~
c': _ _ _.c·
~~
::,~,
;~~
.::.~;.::::. , - - - - - - - - - - - - - - -
~.:" .i-: -,{'-\~: .:_:
,~;~,;. - f . : : " : - - - - - - - - - - - - -
:'>.":.%
+______
:.".:%
+-------.;;:i=------'L:o;.:-:- - -
(~ ~:;;~~-:
..:~.:/.:.,
:: - - - - :..----':
- ...- - - +- - - - + -
:.:
::i'::~.
.....__
/ - /...
_ _ __
".:;.',:, +--------!f~:i-}-··-··---~.:---.......:-:,.:.......
(:::i'::~.
,....•,.
,.-.-,-.-.,.
~>
(~"~;.-'4
.. ~~~~~
[.~~
.-.
...1
:~~:
{,:
1""~'"
(.~
.. ..
.- -..,.: '
..•...: :0
~J:i-
---,-._.i.---:)'i--::.
H
,C
~~i
;
••••
,••, . - , - - -
tq
,"",;'\
,C
~':
~~i
:.:~
\~~
')
L
o
..
,'\',
z:.~:
~~:
~~:
~~:
' '
Note: At the beginning of each month, all securities carrying an investment-grade rating are grouped together to form a
rating cohort. The latest one- and five-year cohorts are formed on July 1, 2007 and January 1, 2002.
~~~~~~~~~~;~:~~~~~~~~~·.·.November 2008
::::: Special Comment ::::: Moody's Credit Policy -- The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Figure 4 (continued) - One-Year (yellow line) and Five-Year (blue line) Investment-Grade Loss Rates
~: U~(.
-,---------------
~:O%
\:\~;~-:.~.
+-------------,
,'..l)',:, + - - - - - - - - - - - - -
:::U);.
+-------------,-
-,---------------
A,(,\\~~(
+-------------
·::.~-r:_:)
+-------------
"p~~_/
. ::.;...,.....7-:::..:..,.......:........................._...:..,
........................... ,.. ~
H
~~::
~::,
:::~:
('J
;'.~
~~i
~~:
:3~
~~:
:3~
~
~(;:
;:~.'
?:
')
;:."";.
:';D'}::'
::
·············,~:;"··~.,·,··t"··~···t"·~.:..!(·,.·4
8 8 ~ 8 '"
~
~~':
~':
~~
;;:
.';:
.';:
(:
.~:,
..
~
:.~.'.:.%
-,---------------
2.,)':';
+------------~
·,.'~.'i~.
+-----------------c,
"lei°'. + - - - - - - - - - - - - -
.,.,)%
+-------------"
:::. W%
~::.:.:~_S(:.:,
-,---------------
+-------------
::.: ..,:% +-----------------'~,
m
~::;
;.~~
:3~
~
;:."";.
:3~
;:."";.
~
,,< :)% - , - - - - - - - - - - - - - - -
'(:0% - , - - - - - - - - - - - - - - -
·,·:to',:, + - - - - - - - - - - - - - - - - - c
::2.(ns + - - - - - - - - - - - - _
+-_ _ _ _ _ _ _ _ _ _ __
J.D% + - - - - - - - - - - - - -
,,1:)%
:~~::
.:;:
~~
:~~
I
~:~
.."
r-)
~
~:~:
,,'
.;~~
~\'
~~
:~~
:'3
(:
.:;:
C·)
'0;'
(:
C·)
'0;'
.:~~
-,---------------
;~•. ~-r:.;)
-,---------------
:), b',:, +-----------;~,~\:,,__---I: .."!~
C':)%
+-------------:
<t.D'::'
+--------------c:,~~
:~:.:2D::.;)
,0-----f"J
:
(:,:::~:';'~~\ +---------;: : :~:C__----
,'1:)%
::.:.:-.{r~~S
+------~,
I.,J. ". .} e
··'··\··'··'··'··'··'··'···'··'··'··1··,··,0','
c;j
;~::
$.:
S~
.."r-:
.~~
.'~:
;~~
~
~~~~~~~~~:;:::~~~~~~~~~·.·.November 2008
=:2.0":-:) + - - - - - - - - - - - - "IU':'. + - - - - - - - - - - - - :::.~;::.:, + - - - - - - - - - - - - -
"
., ...~.' ...
,'~,
::.~.
:'3
~~
.;~~
:.
;:0',·;, + - - - - - - - - - - - - - : :·~.D%
+-------------:-
'0',·;,
+-----------.:.'--
t··
I
:,:'")
:::.D%+·,····,···,·····(···,····,····,····,-···,····,··..,....,-...,....,.....,....;....,....,....,..::,.
) ...
..
<~,
.."r-:
c~
.--:.
~
.~;-.
,~,
c~
.--:
.
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Average Rating Before Impairment
Figure 5 presents the 36-month-average rating before impairment over time as well as the number of newly
impaired securities used to calculate the average ratings. The following observations are noteworthy:
::
The 36-month average rating before impairment for global structured finance rose to Baa2 for the most
recent cohort, one notch up from Baa3 six months prior, and four notches above Ba3 twelve months prior.
Excluding SF COOs, the Other SF category, and 2005 to 2007 vintage US HEL and US RMBS, the
average rating before impairment was two notches lower at Ba1.
::
The average ratings prior to impairment for US HEL, US RMBS, and global COOs remained high at Baa3,
Baa3, and A3, respectively. For global COOs, this represents a 5.5-notch increase from the year-prior
level, while the average ratings for US HEL and US RMBS increased 1.3 and 2.6 notches, respectively,
over the same period.
::
The average ratings before impairment for US ABS excluding HEL, US CMBS, and international structured
finance excluding COOs and Other SF were B3, Caa1, and Caa1, respectively, low in absolute terms and
compared to their historical averages.
Figure 5 - 36-Month-Average Ratings before Impairment (yel/ow line) and Number of
Newly Impaired Securities (blue line)
.~'\r.:~~,----------------, "i~~K:
.',,':,'
mH~
."". '", '':''·''.';S''"
~.~,
,on. & >,<'\!m:,
':'.,;
/\
. I>);
-="
,(
+ - - - - - - - - - - - - - - ; 0 - + '1':.;);
:'-': :1(1~: ~.I ~.r
••••
~~
••
b"~
•••• r.i •••• ' •••• i.' .. r.r•.¥
..
'."
8"~
,~_.
~~~'T
.......... /
(3
~~
'.~:.m/"""·~ ,: '
tq
.....
;
..
,'\',
:-.:":
' '
~~:
~~:
~~:
.C
~~i
~~:
:0
'-'1
:.:~
.,
~':
.,
~':
\~~
\~~
.,',.:,',-------------,!.K;
.'\:.l
,::,x::+-_______+__c:~....";:_::-__+
:':.;
:~:9
........ :
~
.
~'~ +-'.-r'~-
-,.-:f:-;~-:is:- l~ ;:
. -.-.--\-..,-..-.. ,.. ,-.,.-.,-.;....,..-,:..
:~~::
;:.~::
.'~:
"~~
,
-;-,
:'3
c~
:.:.")
~j
..
..
..
,.r-:
~
,.r-:
~
,.r-:
~
~~
;~~
~:
;~~
,::,':'
~>
:.
.,.
:'~;
~~~
'-.~;,
~::.:
t:~~
i~:
~~~
.~
.~
c~
.--:.
~~~
.~~
.C
:.:~
~':
\~J
._)
Note: At the beginning of each month, all securities that become impaired within the next 12-month period are grouped to
form a cohort. The latest 12-month cohort is formed on July 1, 2007. Breaks in the average rating before impairment series
occur when the number of impairments for the cohort is less than or equal to one. Crosses indicate that the average rating
was computed from only one impaired security.
~~~~~~~~~:i:;:~~~~~~~~~·.·.November 2008
":': Special Comment ::':' Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Figure 5 (continued) - 36-Month-Average Ratings before Impairment (yel/ow line)
and Number of Newly Impaired Securities (blue line)
~~~;r:,;+-
_ _ _~:
-------H
\.:.C'--~:"'-':"'-'_ _ _,~i;!-+t
C<: ----'e:: ::::: ':::.,:::,.. ..-,/::';/
+-----------++
..-.':::
·-.. ···..
\'-.-
·······r··'··
···,··r··'··r··'···'""~_~M·
'-~;
~~
'.'
.:~~
------------------- •.•/
~~
,':..-:",
~~
.:~~
':i,(~)
:;,"':n
+ - - - - - - - - - - - - ; ' - - + 7:)
iF;
I.',',
".
-:..-;.
.';:
/-',. + _ - - - - - - - - - - - - +
~:::c~)
/~.:~+_-----------_t .::~.~~
' ' ' · ' ' + - - - - - - - - - - - - H 3C(:
-!--c--," --,.., --,--,' ., .. ' .• --,--; ",-- ,., --,--' --,--,", --, .•..,.., ..,. ','4-1-
(~::;::.:.:
n
.", .. ", ......-............;~......~ ................-..........-....................~ ..... "",;",{ ".;j
:,',
"G, ,
$.:
(?
'"
~S:
~S:
~-!-:
~-!-:
i~',:·o.-,---------------,
z;::;
:----------+
~:C(;
/'...~:+-------c
.-:
I:",:;n+-----~------.C
::>:.:~~+_-----::c::.;" ~.·--'-c····:;;;.·4~:-:. . -._L
l~J+-
_ _ _ _ _ _ _ _ _--,,: . __++
(>-~;J':~ -r"-','-'Y
.~
••••••;-.
•••-•.~.•••.--,C
••••-.'i-'
.....:p;
......"
......'"e.,.~e'-r-+
•••••••••••••••, •.•,•.•.•, •.•,
~~
:.)~
:~~
":J
:~~
":J
~+:
.;~:.:
,'~.
::..~.
~~
~~
~~X:
D
;:;:,+---------oi-.--=-\.,~.-~:----c
I..:;::;::'::
····'····l····'·······r····r·j
B
~~
":J
..,,))
:~~
(:
~~
(:
..,,))
,..f'·
'."
:~:
:s:
:s:
:s:
,,'
~':
.';:
"
(~.:
.';:
-'~;r:,'-,----------------,
ill
;\;;+--------------;,.+
1:·:',::,:·0.+------------,:".4
F*!:~+_--------c__--_+
:::
~*:~~+-------------,--+
:..:-..:-+------------,:~
g:~
.;~~
~~~~~~::::~~::~::~~~~~~·.·.November 2008
~~
..:~~
~
:-;-,
:'?,
:~~::
.';:
'.'
.:~~
~j
h~,C
'0;'
.~~
,.r-:
''';'\
:.:~
~':
\~")
._)
Ie:'
··:::)1 n"
·\·,,····,··\.0::.•
,~',
/ \ : : : " : . : , - - - - - - - - - - - - - - - - - - , -?O
+-------~:-----_+
n'
i:;,
i~:
:"': Special Comment :"" Moody's Credit Policy" The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Rating Action Rates and Large Rating Action Rates
Figure 6 reports 12-month rating action rates and large rating action rates. Figures 7 and 8 further
disaggregate rating actions into downgrades and upgrades, and Figure 9 demonstrates how frequently
downgrades and upgrades have been preceded by watchlist actions in the same direction.6 Key observations
include:
:':'
The 12-month rating action rate in the global structured finance category was 23.7% for the July 2007
cohort, more than twice the rate of 9.6% six months prior and almost four times higher than the historical
level of 6.1 %. The large rating action rate was only a little lower than the overall rating action rate, jumping
to 20.6% from 6.4% six months ago and 3.5% historically. Excluding SF COOs, the Other SF category,
and US HEL and RMBS transactions that closed between 2005 and 2007, the general and large rating
actions rates were much lower at 13.9% and 10.8%, respectively, but still high compared to the past.
::
The global structured finance downgrade rate reached a new historical high of 22.2% for the cohort ending
June 2008. The large downgrade rate followed the same pattern, rising to 20.2% from 5.7% over a sixmonth period. Except for US CMBS, all sectors of structured finance experienced large increases in the
frequency of downgrades.
::
In contrast, the upgrade rate for global structured finance and all the sub-sectors displayed a declining
trend. For the most recent cohort, the overall frequency of upgrades was a low 1.4% and the frequency of
large upgrades was a mere 0.4%.
:':'
Although still low by historical standards, the proportion of downgrades that were placed on review prior to
the rating action rose to 35% for the cohort ending June 2008, up from 22% six months prior. All sectors
experienced an increase in the frequency of reviewed downgrades over the previous 6 months, with the
exceptions of US CMBS and international structured finance excluding COOs and Other SF which were
flat; US ABS excluding HEL and global COOs experienced the largest increases. Meanwhile, the
percentage of reviewed structured finance upgrades continued to decline, falling to 18% for the cohort
ending June 2008 compared to 23% for the cohort ending December 2007.
Moody's also reviewed its uses of watchlist actions on structured finance securities in a Special Comment, "Structured Finance Watchlist Resolutions ..
1992-2003," June 2004.
6
~~~~~~~:::~~~:::~~~~~~~·.·.November 2008
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Figure 6 - 12-month Rating Action Rates (yel/ow line) and 12-month Large (three
notches or more) Rating Action Rates (blue line)
<,,;';'.
~------------
:.";%
+-------------
:.'(;~;'.
+-------------
.,,;,;,. + - - - - - - - - - - - - .::::: + - - - - - - - - - - - - - } .
................. ...
..'
. . . ,. . "'.:......... · · · . 1-
~---~
(:y
l···:·::···,·······::::······:::·::,:·::::·:···::·:·:·::.>/
8;
~.!:
G::
~;~
~"';:
.:~:
<,,;';'.
(~
i';
..
~
;:."";.
5';;
..,........... .
:~:::
+-------------.::
~,~;, ......~
5';;
':,:
'":J
.(~:
t~.;
~~
~:
(:
'-:,:
:s:
,,)
'":J
'0;'
~------------
?:
...............
~:.;
(~
~~;
;,':::%
,~.
?:
..:V-;'::'::'
+-----------------c:j
::(:%
+-------------:
:s:
C:
}~
~?
(:
,,)
'0;'
:~.:
:s:
c.
C:
}~
~------------
G::
c(
:'.:;'~'; ~------------
:
.:)'
:./......: .......... ........}
,;';' I·:-·················;········:::::··,··::··:~······:::·::/
;:."";.
2~:
~;~
~.!:
'"1"'
~"';:
(~
;:."";.
;:."";.
,~;
~
~
i·~·:
c(
?:
~.~;
;.~~
(~
;:."";.
,~;
~
,~.
c(
?:
~~~:':'~. ~------------
~:;(-,~~~-:
+-------------
Z:% +------------~r
,':;'~';
+-----------------,.,;
~ c~~~-:
.,,;,;,. + - - - - - - - - - - - - : / . ? -
J
:;'~';
r.' .. '.·...
8
}~
~~~~~~~:::~!:i:~~~~~f·.November 2008
::.::%
i
,.j·i\.':.:·rf·::·<::~.....
~
:s:
,,-;<••'(•.••••••.,.,.,.,•••""" ••,.,::.,.,.,,:.,.,:::)
~
}~
~
~
:s:
..
(:
.~:,
~
~
:i:" _:f~'<: : ,..F.~·:'\i'-:·:::·:~:".~~·:.;::~:.:/
,;% ...;:...•....,-'-';,. .
$
8
~
:.~::
2:
:'~';
:~:::
(:
\:3
~~
..
.~:,
.';:
~~:
:~:
:~:
~~:
:~:
~~i
(:
~
:':': Special Comment ':':' Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Figure 6 (continued) -12-month Rating Action Rates (yel/ow line) and 12-month
Large (three notches or more) Rating Action Rates (blue line)
Global COOs
Global COOs ex SF COOs
35% ,------------------------------
35%,-----------------------------
30% +------------------------------
30%+-----------------------------
~~~
25%
:!
/<\
: : =:·-----c;f~~
00/0
I····c·.:.~.·.·t····"(·;~·····,·....,:··i·:·::;;:,<r···~~~~~~~~~~
co,
,-,.
i-\~
\;.'
;.'
.;.-:
,y~
if
.;.-:
~~;
,y~
:.,
...
'-:,",
..
~~
+-----------------~----------
~~~~::> .'-. -. :-.:_,
10%
5%
7-iU
00/0
' .' ..'_.'._. . . . . . /.'
/;':',..:. .\. . . . . .'-..'. .:. .t::'\:.-,.....••......
:!I,
·I··~··,··t··,··,··,···,··r-·",--,--;'";-~,-~~~~~~
,-,-..
:.,
...
C
Cohort starting [ate
:":.
:":.
~~
~~
;:;;
other SF
Int'I SF ex COO & other SF
20%,-----------------------------
;:.
:.:...::
12%
8%
c--~----c---~./. <............)</ .....•. <.~.
4%
0%
~:-~,
Cohort starting [ate
20%
16%
-~~;
,I. x. . . ,. •,.;).: . . ,.•. ;•.• : .· : : • .:· ·: ·:::{). . . . . . .,..;. . ./. . . :, . . . . . . . . /
t;,
.-,;
,,-.;
'.
c;
."
.
-,;
T;\
Cohort starting [ate
..
: : I~L
j
::LW
\;.'
-.
,-,-..
;~
.;.-:
....
~
,y~
if
.;.-:
,y~
l~:
...:.,
'-:,",
.;.-:
,y~
Cohort starting [ate
Note: Rating actions include upgrades and downgrades, which are measured on the alpha-numeric (or modified) rating
scale. Rating cohorts are formed each month covering a 12-month period. The latest 12-month cohort is formed on July 1,
2007.
~~~~~~~:::~~::~::~~~~~~·.·November 2008
:':': Special Comment ':':' Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Figure 7 - 12-month Downgrade Rates (yel/ow line) and 12-month Large (three
notches or more) Downgrade Rates (blue line)
<,,;';'. - , - - - - - - - - - - - - - - -
:.";%
+-------------
~ ~;;~~-:
:.'(;~;'.
}
.:~}~~~ +-_ _ _ _ _ _ _ _ _ _~J
{
.:::.:.~.
c~~~-:
;;.
,.>. ,-. <.> . . . :.: .........: .:::i~
r.;:;:-:;·.·••:.:.::.:-:-:-:.:.;••••::.,;•.•:_;:.;:::::::-.-;:-:-:~,.;:..;:;:;-;-::.:;;-./
g:~
.;~~
.,:,,;,.
~: ....
~:~
'"
~\'
..r-)
~~
,~
Si
:~~
,-~.
:~~:
S
~~
.'~:
,~:
+-------------
.,,;,;,. + - - - - - - - - - - - - :(:%
~:; -: ~-:
+------------_.
: :~_.~-" ---:/0-
+--------:---;:
. . . ....
:"C7.";.:
.......;
.........
-
'oo?
:.Y~~~ :"::;';-;:':":;:':;:-;-;-;-:;":"-':"""!(c'.;-;:~;:-:::;";::'-'''-l;-;-;:'"'':''4''-;-';''~~~=~
~~
~:
(:
:s:
C·)
'0;'
-,---------------
:s:
C:
}~
~?
(:
C·)
'0;'
:'l% - , - - - - - - - - - - - - - - -
": :.?!:~~
:~
'::.Y;(·
;~ tlJ1
,~N_
. .: . - -
1:
'~\':'~.
"
:;;.0;:7.;~:.-.:.:••••;._....
•.
•.•.-,",....--
....
:-.':':'~. l.l_r_T_'_T_~_:(::~;·:·.~::::-::<:::·:·:··:·~:~;'"-,··i~'~~~=~
~~';
:~:
~.;
~~i
:'~';
:~:
~'.)
(~
,~;
2~:
~:.;
?:
~~;
(~
,~.
c(
";-;
:~::
?:
:'3
.';:
~V;'::::'
.;~
'};",:,
·~:u:.~.
t
: .~-.
+-- - - - - - - - - - - - - - - - ; . :. '
,....,,,~,
::.,. +-_ _ _ _ _ _ _ _ _ _ _----:l:_
::i'::~.
,.···c··· ••• r
G::
. ....~....,
,C:,. t:)q·
"1"'
~"::,
i'~':
('J
\.";.
:3~
~~~~~~~:::~~:::::~~~~~~·.·.November 2008
~
:3~
'r'
. . . . . . . . . . . . . . . .l
:~:::
t:3
~~:
~~i
2.:
:~:
S~
.;~~
S~
.;~~
,'~,
:~~:
:'3
.';:
~~
.;~~
::,~,
:'3
~~~:.%
-,---------------
"K:%
+-------------
~:;c;~~·:
+-------------
:-'(;~;'.
+---------------;:'
:.l
.,,;,;(. + - - - - - - - - - - - - + .
:.}:;(.
.;;.... ;{.:
.......
,.......'....~: ........:...... . •..................................... ;.;;:.:i'.
·,··~.:.;··T··iC:;,.
...:.~:::
~:~
'"
,:~:
~'"
~'"
";'
!~
!~
~::
,-r>:
'c'
.;~~
~j
:$
";'
!~
,-r>:
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
b
.;~~
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Figure 7 (continued) -12-month Downgrade Rates (yel/ow line) and 12-month Large
(three notches or more) Downgrade Rates (blue line)
Global COOs
35% ,------------------------------
Global COOs e x SF COOs
35% ,-----------------------------
30% +-----------------------------~
30%
~~~ +------------------./
25%
...
}::-
'"
:~
15%;:/'\\ .
10%
5%
00/0
-1 •• I ••
~.'
....'!
\.....
....... :~.--··~·--.....
•• l •• ,•• ' ••
:'1
(:
~'~
r ••
i.'.:·.r~:· ,. . . .-..-. . . .
.'.
c;
..
".
,-':
;):'
:~3
")
'\"'::;.:;:{:..:::'
..
:.r-
i;'
,-)
-=-;.
:~3
.:;:
")
10%
5%
00/0
.\.,.:,....
1
;.
~{.-,.----\\~--:~.f
.);.;:..:....._.,.;......;;.
·····1·-'i···-·-·-·-···-·-·
:.,
...
Cohort starting [ate
Int'I SF ex COO & other SF
12% ,------------------------------
other SF
...),--------------------------
8%
6%
4%
2%-----~-L-----=..::::::::.~------~,
0% J., .. ,., .. ,. . . ,.. ,.. ,., .. S.,., . ,
. /::··.:?:·/::····:::..:::.•............)
:~~
':,
C
~,?
1"--.
):'
C;
:))
,~
;
(:
.~)
c::.:
,~ ;
")
Cohort starting [ate
:0
..
:S'
::")
(:
.~)
")
C
:
I::
.:
.....)
.......:
+ - - - - - ' - - - - ' : · 7.~
Cohort starting [ate
10%
l
...
~~~ +----------------:;"'. .;:
~i"
(:~
".
i
+----------------------------+-----------------------------.
::~
~~~ :1============================:;:
1:~ -.--~~~~~~~~~~~~~~~~~~~~~~~~:-:.i·
+1'.-.
4%
2%
•.•
,-: .. :
0% ){..;. ,.....,.:l,.....),~:.. 1
../. f"'Y'(':~'~'~;""";':'::~'~:::::'r"" ..~
..
'0
;:.
~~;
';:.
~~;
)~
(
"
'-:'-,
'.
~~
Cohort starting [ate
Note: Downgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month
covering a 12-month period. The latest 12-month cohort is formed on July 1, 2007.
~~~~~~~:::~~~:~::~~~~~~·.·November 2008
:':': Special Comment ::':: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Figure 8 -12-month Upgrade Rates (yel/ow line) and 12-month Large (three notches
or more) Upgrade Rates (blue line)
~::';,~:,t:(;:l ::.t:'~J(;t:.x·:::-=.: f:;):<:):G-::~ (:)X ~~F
"";'
-,------------------
,t,,~+-------c
:::
--
------~,c'-.
.;/~~.
,:::;~'~f:~ 'Q:~.. '{~'? 'V~?~b:)(:::)
.
;:.~::
~:~
.'~~
..
,r-)
~
~+:
,,.-.~
,-~.
.;~~
}~V
~
: : , - i - - - - - - - . : - ,- - : -
;:~~y~
"-------_."
;~~
C':
}~
~~
:~>:
~-:~
:s:
..
(:
.~:,
~
g;
'"
~?
.:~~
;;:
'0;'
(:
,,)
t~
}~
:s:
i";'~~:' +----------~:
"
~::':.}~~(
+-:----':-:-----------:-;-
,~~,
i~
c_
,~.:
C':
~-!-:
.i~
~~
'~.'
(:
,,)
.--:.
-:s:
t:~
.i~
.:
·1··~S~,
1
c; 0'%.
~;~~ .
~B', +--- ~/;-,-----------c:~~/::\,3';',
S~
c;::()s. Ot:)::)~-
U"s. ;·fL .'S,. ~~.....~~=-:~;,
. .'"
':
,"S'. + - - - - - - - - - -.......(: \-
~n
;.~~
:-.....
(:
.~:,
~~:
._)
~~i
-,
,;j
(~
,~;
-,
,;j
(~
;:."";.
,~;
~
:,
:..:)
:3~
:,SI'n.
';"0 - , - - - - - - - - - - - - - - - - - -
+-------------
:j";:;
+------------c:~:-
?',;: + - - - - - - - - - - - - -
)%
+----------------c:-:-
:::':,~
,"
-:s:
~~~~~~~:::~~::~:;~~~~~~·.·.November 2008
,'1% - , - - - - - - - - - - - - - - - -
gs
.-,
(:
(:
.~:,
.~:,
._)
._)
~~':
,~.:
::~,
~s:
~-!-:
.--:
.
,,)
~:;
(~::
(:
;:."";.
?:
~
~~i
:':': Special Comment ::':: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Figure 8 (continued) - 12-month Upgrade Rates (yel/ow line) and 12-month Large
(three notches or more) Upgrade Rates (blue line)
Global COOs
14% - , - - - - - - - - - - - - - - - - -
12%
;:
10%
~ ~~------------
":.------------
8%
~: i- - - - - - - - - - - -
6%
-------------
Global COOs ex SF COOs
14% - , - - - - - - - - - - - - - - - - -
:~:------------
12%
.....
10%
- - - - - - - - - - - -
8%
6%
4%
4%
2%
••
0% ,----).;. t-----(--~-----t
.;L.;:
::-,::~:~
.'
.'
.. . ......
0::G~.-./:;::::::::::·:::·::::;:·;-;·:."-r;;:;-~;:·-.-','.'""c".~;.,~~T";
g~
\;.'
~,;
~~
'.
~,;
'.'
.::
,y~
•
';-.- - - - - - - - - - - - - - - c
2%)-·- ••
:;
--c----------c---c/:::.:
",
.
......... <:. . . . . ;' . .
:
0% -1 .. I.~.~.' .. T.:: ...,.. r:,:·::,~::;S'.,.:.:·:·::-:)·:·:":.:................."-""
.-.-)
.....,',
~:
:,';:
:-;.;
:;..\
;:.;:'
:(;
::~
"
'-:'-,
Cohort Starting Date
.-.-)
'.'
,-;
:;..\
,-:~
C·)
:(;
Cohort Starting Date
Int'I SFex COO & other SF
other SF
8%,---------------
5%
7%+-~-------------
4%
6%
3%
2%
1%
;J:
,-;
I..!',:
c:
'-'
~:.
;(,
~~
,-;
Cohort Starting Date
"-;
c:
0%
I..!',:
c:
(;
~:.
;(,
I::
.':.
-
..
:s'
):'
:))
(":
::?
..
..
~~ ~
\
\
.::
..
:s'
Cohort Starting Date
Note: Upgrades are measured on the alpha-numeric (or modified) rating scale, Rating cohorts are formed each month
covering a 12-month period, The latest 12-month cohort is formed on July 1, 2007,
~~~~~~~:::~~~:l~~~~~f·November 2008
••••• Special Comment ••••• Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Figure 9 - Percentages of Downgrades (yel/ow line) and Upgrades (blue line)
Preceded by Watchlist Actions in the Same Direction
'ln~::;~~
-,---------------
,
-;-,
:'3
:-.:~
~~::
C.'
",
?:
?:
'-"1
,;j
.~.
I};,f. +".:;...;,
••:.'-,-::'~~~~~~~~~~~
~:~:
~\'
-:..-;,
:~~:
.'~:
..,
:~:
~~
(:
~~
.'~:
.'~:
S'(~
",:.'
((~:
:'-.
,~;
,;j
",
?:
,~;
.~:,
..
~'.)
(~
~
~n
,"",;'\
(:
.~:,
..
~
~-:~
(:
.~:,
..
~
gs
(:
.~:,
..
~
.",
(:
.~:,
..
~
[~
(:
.~:,
..
~
B
(:
~:)
(:
.~:,
..
.~:,
..
~
~
;:'~.'
:~~::
...
'~:
;~~:
::..-;
.';~
.';~
.';~
.';~
.';~
.';~
~: ....
~~
:,:'")
::..-;
.;~~
'0;'
.';~
ttr:::tr"November 2008 ;:::; Special Comment:::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
:::::::::::::::::::::::: ...
,.r-:
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Figure 9 (continued) - Percentages of Downgrades (yel/ow line) and Upgrades (blue
line) Preceded by Watchlist Actions in the Same Direction
Global COOs
Global COOs ex SF COOs
100% ,-----------------------------
100%,----------------------------
:::/::cJ~J~
W%
00/0
:I_(:;-_·f_\:c____--;{::-'----_________
,·.·.~.·.·t·.·.~.·."{.·.·t·.·.;.·.·t· · i· :·;~,- J· i·... ;:~:;·-·~}~~~~~~
t ••••i ....
'-;;.
~~;
::-:.
'-;;.
~~;
,--',
'-:'-,
..
~- '
'-:'-,
...
...
.;.-:
.;.-:
;-~~
.;.-:
,y~
,y~
,y~
80%
+--------
~~
~.~.
.~~
"'::-------c--:~-L-
::: +---------:.·---c;!J~~..r\j1
+----11 ·L::. J
20%
0% -,.. ,.q., ..,..,.q., ..,..,~.. """",,~~~~~~~
'-'-.,
L!-.•
"-.;
".
:~~.:
(:~
:~(
c;
c;
c;
(~ .
(~ .
-;~:
~~;
-;~:
Cohort starting [ate
-;~:
-;~:
..
..
..
..
Cohort starting [ate
other SF
Int'I SF ex COO & other SF
100%
100%
80%
80%
60%
60%
40%
40%
20%
20%
0%
0%
:~~
c
~,?
:~~
C
~,?
..
::.
::.
\
Cohort starting [ate
..
\
~~-~
\
.'~,
c
~,?
::;-:
:.:;-:
;.;:
,-;,
c
c
-~~:
-~~:
-~~:
-~~:
-~~:
-~~:
I!-.,
c
c
-~~:
-~~:
Cohort starting [ate
Note: At the beginning of each month, all securities that experienced downgrades (upgrades) within the next 12-month
period are grouped together to form a rating cohort. The latest 12-month cohort is formed on July 1, 2007. Gaps in the data
indicate that there were no downgrades (upgrades) during that time period.
~~~~~~~:::~~:i;::~~~~~~·.·November 2008
::::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Appendix: Description of Data Sample and Glossary
The data sample used in this report includes all public, 144A, and private tranches with a published Moody's
long-term global debt rating among global asset-backed securities (ABS), commercial and residential
mortgage-backed securities (CMBS and RMBS), collateralized debt obligations (COOs), and other structured
finance, including asset backed commercial paper (ABCP), structured investment vehicles (SIVs), structured
covered bonds, catastrophe bonds, and derivative product companies. Provisional ratings, credit estimates or
evaluations, short-term ratings, and national scale ratings are not included. The following types of securities
are excluded from the definition of global structured finance and therefore are not included in the data sample:
repackaged securities, structured notes, and other credit derivatives which are basically pass-throughs of the
rating of another entity.
This data set is an expansion of the data set that was used in prior structured finance performance studies. 7
Unlike the data set from previous years, this data sample:
::
Includes tranches wrapped by financial guarantors, government agencies, and government sponsored
enterprises (GSEs);
::
Includes interest-only (10) and residual tranches;
:.:.
Includes some transactions outside of the four major sectors (ABS, COO, CMBS, RMBS) of structured finance,
such as ABCP, SIVs, structured covered bonds, catastrophe bonds and derivative product companies;
:':'
Does not collapse tranches with the same rating from the same deal, i.e. all pari passu tranches are
counted in the data sample. The exceptions to this are notes with the same rating issued out of the same
program for ABCP, SIVs and structured covered bonds, in which case only the rating of the program and
not each individual security is counted.
The data used to create this report are commercially available via Moody's Structured Finance Default Risk service.
For more information, please email DefaultResearch@moodys.com.
Glossary
Material Impairment
Structured finance securities are defined as being in material impairment if they have:
:':'
Sustained a payment shortfall that remained uncured, or
':':
Been downgraded to Ca or C.
Prepayment-related and AFC-related interest shortfalls are not considered to be material impairments, but
PIKing tranches are. Explicit principal write-downs are included whereas implicit principal write-downs or
under-collateralizations are not.
The impairment status of a security may change as it goes from cured (i.e. all outstanding shortfalls and losses
were repaid in full) to uncured (i.e. positive interest shortfalls or principal losses outstanding), or vice versa. If
any securities rated Ca or C but not in payment shortfall are upgraded, they are considered to be no longer in
material impairment. Securities rated Ca or C that were not upgraded are in material impairment even if their
payment shortfalls have been cured. Finally, securities with very minor shortfalls or losses are excluded.
The expanded data sample was first introduced in our 2007 structured finance rating transitions studies that were published this year. The data sampl
in this study was extracted following similar guidelines.
7
~~~~~!:i:~~::~::~~~~f·.November 2008
:':': Special Comment ':':' Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Payment Shortfall
Structured finance securities are defined as being in payment shortfall (previously called "payment default") if
they have suffered either one of the following:
':':
Interest shortfall
:':'
Principal write-down.
Principal Impairment
This refers to materially impaired securities that have suffered principal write-downs or principal losses, or
have been downgraded to Ca or C even though a principal write-down or loss has not yet been observed. In
particular, if a security had experienced principal write-down/loss or was downgraded to Ca or C, it is called a
principal impairment regardless of whether it had experienced interest shortfalls.
Interest Impairment
This refers to materially impaired securities that have experienced only interest shortfalls, no principal losses,
and were not downgraded to Ca or C.
Investment-Grade (IG) and Speculative-Grade (SG) Ratings
Investment-grade ratings refer to Aaa, Aa1, Aa2, Aa3, A1, A2, A3, Baa1, Baa2, and Baa3. Below investmentgrade or speculative-grade ratings refer to Ba1, Ba2, Ba3, B1, B2, B3, Caa1, Caa2, Caa3, Ca, and C.
Loss Severity or LGD
The LGD rate of an impaired structured finance security is measured by the sum of the present values of net
losses, including both interest shortfalls and principal losses, discounted by the security's coupon rate and
expressed as a percentage of a given principal balance such as the principal balance at origination, at the
impairment date, or at any given cohort date.
Accuracy Ratio (AR)
An accuracy ratio based on the loss experience of structured finance securities (or an accuracy ratio adjusted
for loss-given-default, or LGD) is the ratio of the area between a loss-based cumulative accuracy profile (CAP)
curve and the 45-degree line to the maximum possible area above the 45-degree line. The loss-based CAP
curve (or a CAP curve adjusted for LGD) plots, for each rating category, the proportion of the losses of all
impaired securities accounted for by securities with the same or lower rating against the proportion of all
securities in the sample population with the same or lower rating. 8
To calculate accuracy ratios, rating cohorts are formed for each calendar month over the study period so that
all outstanding securities in the sample population are grouped together by their alpha-numeric ratings at the
beginning of the month. For each monthly rating cohort, we determine the number of securities that became
impaired within one or five years of the cohort formation date and their loss severity rates as a percentage of
the cohort date balance. Cumulative shares of securities rank-ordered by rating are calculated for the
universe of all securities and the universe of impaired securities, respectively, and based on this, a CAP curve
is plotted. Note that only LGD for principal impaired securities are used in the calculation. Please see the
definition of LGD for further details.
The CAP curve adjusted for LGD is also known as a "power curve" because it shows how effective a rating
system is at differentiating between securities that have sustained high losses from securities that have
sustained low or no losses. The metric is defined relative to the distribution of ratings in the population. The
accuracy ratio is an effective way to summarize the CAP curve into a single number. The accuracy ratio is
8
For an illustration of the CAP curve adjusted for LGD, see "Default & Loss Rates of Structured Finance Securities: 1993-2003," Moody's Special
Comment, September 2004.
~~~~~!:i:~~~:::~~~~~f·.November 2008
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
zero if the CAP curve collapses to the 45-degree line, suggesting that all impaired securities are randomly
distributed throughout the population without regard to rating.
Investment-Grade Loss Rate
The one-year investment-grade loss rate is calculated as follows. First, for a given cohort, we calculate the
LGD as a share of the tranche balance at the cohort date for each security that carried an investment-grade
rating at the cohort formation date and became impaired within a 12-month period after the cohort date. We
then take the sum of these LGD rates and divide by the total number of investment-grade securities
outstanding as of the cohort formation date. Note that the LGD rate is not weighted by the total dollar volume
of outstanding securities in the cohort. Also note that only LGD for principal impaired securities are used in the
calculation. The five-year investment-grade loss rate is calculated similarly.
Average Rating Before Impairment
The rating of an impaired security is measured every month for 36 months prior to impairment. These 36 rating
measurements are averaged together to create one representative number for each impaired security. For a
particular cohort, the average rating before impairment is the weighted average of these average ratings for
each security that became impaired within 12 months after the cohort formation date, where the weight for
each security is the LGD rate of the tranche as a share of its original balance. This weighting scheme will
place greater emphasis on the average ratings of impaired tranches with higher LGD over impaired tranches
with lower LGD. 9 Note that only LGD for principal impaired securities are used in the calculation.
Rating Action Rate
The rating action rate is defined as the number of securities that experienced a rating change within a year
after cohort formation divided by the total number of securities outstanding at the cohort formation date (the
beginning of each month). Rating changes are measured on the alpha-numeric (or modified) rating scale and
are based on comparing the rating at the beginning and end of the time period under consideration. However,
if a rating was withdrawn by the end of the time period, then the rating prior to withdrawal is used as the end
rating. Note that a security will only be counted if it was outstanding as of the cohort formation date.
Downgrade rates and upgrade rates are measured similarly based on downgrade rating actions and upgrade
rating actions, respectively.
Large Rating Action Rate
A large ration action is said to occur if a rating action (or cumulative rating actions) cause(s) a security's rating
to change by three or more notches within a year after cohort formation. The large rating action rate is the
number of such securities divided by the total number of securities outstanding at the cohort formation date.
Large downgrade rates and large upgrade rates are measured similarly based on large downgrade rating
actions and large upgrade rating actions, respectively.
Percentage of Downgrades (Upgrades) Preceded by Watchlist Actions
in the Same Direction
This metric is defined as the total number of downgraded (upgraded) securities that were placed on the
watch list in the same direction before they were downgraded (upgraded), divided by the total number of
securities that were downgraded (upgraded) within 12 months after the cohort formation date.
We began using LGD rates as weights in computing an average rating before impairment in the full-year 2005 structured finance performance report ..
Ideally, LGD rates should be calculated as a percentage of the principal balance outstanding for each month in the 36 months prior to the impairment,
and ratings should then be weighted by these monthly LGD rates. Practically, however, it does not make any material difference in the average rating
number whether we use LGD rates as a share of impairment-date balance, original balance, or monthly principal outstanding. Since the monthly LGD:
rates are very time consuming to compute due to amortization, we use the LGD rate as a share of original balance as the weight variable.
.
9
~~~~~!:i:~!:i:~~~~~f·.November 2008
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
ABS ex HEL
ABS stands for asset-backed securities. This structured finance sector includes securities backed by asset
types such as auto loans, credit card receivables, student loans, and manufactured housing loans, and nontraditional asset types such as mutual fund fees, tax liens, tobacco settlement payments, and intellectual
property. Home equity loans (HEL) are excluded from this sector.
HEL
The home equity loan or HEL sector includes securities backed by subprime (B&C) mortgage loans, home
improvement loans, high loan-to-value (high LTV) loans, home equity lines of credit (HELOCs), and closedend second-lien loans, as well as net interest margin (N 1M) securitizations. It does not include securities
backed by Alt-A mortgages, which are included in the RMBS sector.
Prior to 1998, RMBS collateral was generally defined as first-lien residential mortgages, regardless of the
credit quality of the borrower. HEL collateral generally included junior liens such as HELOCs or closed-end
seconds. However, as subprime lending became more prevalent, the market shifted its definition such that
HEL encompassed subprime first-lien residential mortgages while RMBS included first-lien mortgages made to
higher quality borrowers. Since 1998, a deal classified as RMBS by Moody's is generally backed by prime or
Alt-A quality first-lien residential mortgages, while a deal classified as HEL is generally backed by subprime
first-lien mortgages or junior liens. Therefore, a subprime deal which would be classified as HEL today may
have been classified as RMBS in the past.
RMBS
RMBS stands for residential mortgage-backed securities. The vast majority of these securities are backed by
first-lien prime mortgages or by Alt-A mortgages.
CMBS
CMBS stands for commercial mortgage-backed securities. Commercial real estate (CRE) COOs, where 70%
or more of the collateral is comprised of CRE loans, are classified as CM BS. If the collateral backing the
transaction contains less than 70% CRE loans, then the deal is classified as a COO.
COOs
COOs stand for collateralized debt obligations. Derivative securities such as structured notes and repackaged
securities are not considered to be part of this sector. Commercial real estate (CRE) COOs, where 70% or
more of the collateral is comprised of CRE loans, are classified as CMBS. If the collateral backing the
transaction contains less than 70% CRE loans, then the deal is classified as a COO.
Other Structured Finance
Other structured finance consists of structured finance securities not categorized in the five major sectors
(ABS, HEL, COO, CMBS, and RMBS) including asset-backed commercial paper (ABCP) programs, structured
investment vehicles (SIVs), structured covered bonds, insurance-linked securities such as catastrophe bonds,
and derivative product companies. However, notes carrying only short-term ratings such as commercial paper
are excluded.
Global Structured Finance
Global structured finance captures securities issued around the world in the five major sectors - ABS, HEL,
COO, CMBS, and RMBS - and in the other structured finance category.
~~~~~!:i:~~::~::~~~~f·.November 2008
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
U.S. Structured Finance
u.s. structured finance securities are denominated in U.S. dollars and issued in the U.S. market or
denominated in Canadian dollars and issued in Canada. In cases where the source of the underlying collateral
and the denomination of the securities crossed multiple countries/regions, deals are classified by the location
at which they are monitored.
Inti SF ex COO and Other SF
This refers to securities that are not denominated in U.S. dollars and issued in the U.S. market and not
denominated in Canadian dollars and issued in Canada. The majority of the securities in this sector are issued
in Europe, the Middle East, and Africa (EMEA); the rest are issued in the Asia Pacific region and Latin
America. COOs and Other SF are excluded.
~~~~~!:i:~~:::::~~~~f·.November 2008
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Moody's Related Research
Special Comments:
,:,:
The Performance of Structured Finance Ratings: Full-Year 2007 Report, July 2008 (110179)
:,:,
The Performance of Structured Finance Ratings: Mid-Year 2007 Report, October 2007 (105390)
,:,:
The Performance of Structured Finance Ratings: Full-Year 2006 Report, May 2007 (103017)
:,:,
The Performance of Structured Finance Ratings: Mid-Year 2006 Report, September 2006 (99034)
':':
The Performance of Structured Finance Ratings: Full-Year 2005 Report, May 2006 (97346)
:':'
The Performance of Structured Finance Ratings: Mid-Year 2005 Report, September 2005 (94463)
':':
Default & Loss Rates of Structured Finance Securities: 1993-2007, July 2008 (109707)
:':'
Measuring Loss-Given-Default for Structured Finance Securities: An Update, December 2006 (101284)
':':
Structured Finance Rating Transitions: 1983-2007, February 2008 (107444)
:':'
Japanese Structured Finance Rating Transitions: 1994-2007, March 2008 (107833/ SF125229)
':::
Asia-Pacific (ex-Japan) Structured Finance Rating Transitions: 1990-2007, March 2008 (107947)
::
EMEA Structured Finance Rating Transitions: 1988-2007, March 2008 (107977)
':::
Deal Sponsor and Credit Risk of U,S, ABS and MBS Securities, December 2006 (100872)
,:,:
The Relationship between Par Coupon Spreads and Credit Ratings in US Structured Finance, December
2005 (95494)
':':
Structured Finance Watchlist Resolutions: 1992-2003, June 2004 (87305)
:':'
The Performance of Moody's Corporate Debt Ratings: September 2008 Quarterly Update, October 2008
(112077)
:':'
Measuring the Performance of Corporate Bond Ratings, April 2003 (77916)
':':
Guide to Moody's Default Research: October 2008 Update, November 2008 (112205)
To access any of these reports, click on the entry above, Note that these references are current as of the date of publication
of this report and that more recent reports may be available, All research may not be available to all clients,
~~~~~!:i:~~~:~::~~~~f',November 2008
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Report Number: 112347
Julia Tung
Geraldine Kim
Cassina Brooks
© Copyright 2008, Moody's Investors Service, Inc. and/or its licensors and affiliates including Moody's Assurance Company, Inc. (together, "MOODY'S"). All rights
reserved. ALL INFORMATION CONTAINED HEREIN IS PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR
OTHERWISE REPRODUCED, REPACKAGED, FURTHER TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED
FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN WHOLE OR IN PART, IN ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER, BY ANY
PERSON WITHOUT MOODY'S PRIOR WRITTEN CONSENT. All information contained herein is obtained by MOODY'S from sources believed by it to be accurate
and reliable. Because of the possibility of human or mechanical error as well as other factors, however, such information is provided "as is" without warranty of any
kind and MOODY'S, in particular, makes no representation or warranty, express or implied, as to the accuracy, timeliness, completeness, merchantability or fitness
for any particular purpose of any such information. Under no circumstances shall MOODY'S have any liability to any person or entity for (a) any loss or damage in
whole or in part caused by, resulting from, or relating to, any error (negligent or otherwise) or other circumstance or contingency within or outside the control of
MOODY'S or any of its directors, officers, employees or agents in connection with the procurement, collection, compilation, analysis, interpretation, communication,
publication or delivery of any such information, or (b) any direct, indirect, special, consequential, compensatory or incidental damages whatsoever (including without
limitation, lost profits), even if MOODY'S is advised in advance of the possibility of such damages, resulting from the use of or inability to use, any such information.
The credit ratings and financial reporting analysis observations, if any, constituting part of the information contained herein are, and must be construed solely as,
statements of opinion and not statements of fact or recommendations to purchase, sell or hold any securities. NO WARRANTY, EXPRESS OR IMPLIED, AS TO
THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR
OTHER OPINION OR INFORMATION IS GIVEN OR MADE BY MOODY'S IN ANY FORM OR MANNER WHATSOEVER. Each rating or other opinion must be
weighed solely as one factor in any investment decision made by or on behalf of any user of the information contained herein, and each such user must accordingly
make its own study and evaluation of each security and of each issuer and guarantor of, and each provider of credit support for, each security that it may consider
purchasing, holding or selling.
MOODY'S hereby discloses that most issuers of debt securities (including corporate and municipal bonds, debentures, notes and commercial paper) and
preferred stock rated by MOODY'S have, prior to assignment of any rating, agreed to pay to MOODY'S for appraisal and rating services rendered by it fees ranging
from $1,500 to approximately $2,400,000. Moody's Corporation (MCa) and its wholly-owned credit rating agency subsidiary, Moody's Investors Service (MIS), also
maintain policies and procedures to address the independence of MIS's ratings and rating processes. Information regarding certain affiliations that may exist
between directors of MCa and rated entities, and between entities who hold ratings from MIS and have also publicly reported to the SEC an ownership interest in
MCa of more than 5%, is posted annually on Moody's website at www.moodys.com under the heading "Shareholder Relations - Corporate Governance - Director
and Shareholder Affiliation Policy."
Moody's Investors Service
~~~~~!:i:~~::~:;~~~~f'November 2008
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2008 Report
Special Comment
August 2009
Table of Contents:
Highlights
Introduction
4
Accuracy Ratios
5
Investment-Grade Loss Rates
7
Average Rating Before Impairment
9
Rating Action Rates and Large Rating Action
Rates
11
Appendix: Description of Data Sample and
Glossary
20
Moody's Related Research
24
The Performance of
Structured Finance Ratings:
Full-Year 2008 Report
Highlights
This Special Comment updates Moody's structured finance rating performance
metrics as of December 2008. The highlights of this report are:
Overall, 12,666 structured finance securities became impaired in 2008:
17 in US ABS, excluding HEL, 97 in US CMBS, 6,519 in US HEL, 3,174
in US RMBS, 2,825 in global COOs, 16 in the international structured
finance sector excluding COOs and SIVs and Other SF, and 18 in SIVs
and Other SF. Of these, 12,425 were principal impairments
(experienced principal losses or were downgraded to Ca or C), while the
remaining 241 were interest impairments (experienced interest shortfalls
only).
Analyst Contacts:
New York
1.212.553.1653
Julia Tung
Vice President - Senior Credit Officer
Nicolas Weill
:':
For global structured finance, the one-year accuracy ratio was flat at
60.2% relative to its level six months ago, but declined from its level of
75.7% from 12 months ago (see Figure 1( However, excluding the
most troubled sectors - structured finance COOs (SF COOs), the SIVs
and Other SF category, and US HEL and RMBS securitized between
2005 and 2007 - the one-year accuracy ratio was higher at 74.2%.
:::
The five-year accuracy ratio, which measures the performance of
ratings that were outstanding five years ago and hence, does not
incorporate the performance of the more recent poorly performing
vintages, was 73.1 %, 4.4 percentage points lower than its level of
77.5% six months ago.
:::
The one-year investment-grade loss rate increased to 7.8% for the
cohort ending December 2008, an almost 60% increase from its six
months-prior rate.
Group Managing Director - Chief Credit Officer
1 These performance metrics should be interpreted with caution. Some statistics are based on small samples, as the number of impairments in any
given year and any given sector is often small. Individual performance metrics may also be driven by highly correlated collateral performance
underlying multiple securitizations, particularly within certain sectors. Moreover, variations in rating performance over time may reflect either changes
in the quality of the rating process or changes in the environment that make losses more or less difficult to predict or make collateral performance
more volatile, thus resulting in either higher rating accuracy and lower rating stability at one time than another, or vice versa.
Moody's Investors Service
The Performance of Structured Finance Ratings: Full-Year 2008 Report
:,:,
The average rating during the three years prior to impairment for all impairments occurring in 2008 was
unchanged at Baa2. This was the same as the average rating for the cohort ending six months earlier,
and one notch above the level for the cohort ending twelve months earlier.
':':
The one-year rating action rate rose to 36.2%, a 50% increase from the rate of 23.7% six months prior and
an almost four-fold increase from 9.6% a year ago. Most of the rating actions were large actions, i.e.
changes of three notches or more, as the large rating action rate of 31.1 % was only a little smaller than the
overall rate. Both increases were caused by growth in the number of downgrades, which was seen across
all sectors of structured finance.
::':
US HEL, US RMBS, and global COOs all continued to experienced declines in their accuracy ratios and
increases in their one-year investment-grade loss rates (Figure 2). In contrast, US ABS, excluding HEL,
maintained 96% one-year accuracy ratio and 0% investment-grade loss rate. US CMBS also displayed a
near-zero one-year investment-grade loss rate.
Figure 1: Summary of Global Structured Finance Rating Performance as of December 2008 2
5-Year
Accuracy
Ratio
1-Year
InvestmentGrade Loss
Rate
36-Month
Average
Rating Before
Impairment
1-Year
Rating
Action
Rate
1-Year
Large
Rating
Action
Rate
60.2%
73.1%
7.80%
Baa2
36.2%
31.1%
6,645
60.3%
77.5%
4.92%
Baa2
23.7%
20.6%
2,141
75.7%
79.7%
1.08%
Baa3
9.6%
6.4%
193
90.3%
76.5%
0.05%
Ba3
4.6%
2.1%
1,001
67.5%
82.5%
1.23%
Baa2
8.7%
6.0%
Number of
New Impairments over
Prior 12
Months
1-Year
Accuracy
Ratio
December 2008
12,666
June 2008
December 2007
Cohort Ending Date
Global Structured Finance
June 2007
Average (1993-Most Recent)
Global Structured Finance excl SF COOs, SIV and Other SF, and '05-'07 vintage US HEL 8: RMBS
December 2008
878
74.2%
82.6%
0.64%
Baa3
13.4%
9.6%
June 2008
427
83.0%
81.3%
0.27%
Ba1
10.0%
6.9%
December 2007
219
87.0%
81.2%
0.12%
Ba2
5.9%
2.6%
June 2007
123
90.5%
77.2%
0.05%
B1
6.5%
2.8%
Average (1993-Most Recent)
139
86.4%
83.4%
0.12%
Ba1
5.5%
2.8%
2 A glossary appears at the end of this report. The number of impairments for historical cohorts is subject to revision during each update as
payment shortfalls can be cured and past remittance or trustee reports may be revised. In addition, consistent with Moody's annual default
and loss study, Moody's now derives loss rates using loss-given-default (LGO) from principal impaired securities alone. The historical
average of the number of new impairments over the prior 12 months is calculated as the total number of newly impaired tranches divided by
the number of years in the sample period, and has been rounded to the nearest integer unless rounding results in zero.
~~~~~~~~~:::;:~~~~~~~~~·.·.AU9Ust 2009
':':' Special Comment :':': Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Figure 2: Summary of Structured Finance Rating Performance by Sector as of December 2008
Number of
New Impairments over
Prior 12
Months
1-Year
Accuracy
Ratio
5-Year
Accuracy
Ratio
1-Year
InvestmentGrade Loss
Rate
36-Month
Average
Rating Before
Impairment
1-Year
Rating
Action
Rate
1-Year
Large
Rating
Action
Rate
.................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
~
Cohorts Ending December 2008
US ABS ex HEL
US CMBS
US HEL (includes subprime)
excl ·05-·07 vintages
US RMBS (includes Alt-A)
excl ·05-·07 vintages
Global COOs
17
96.0%
85.4%
0.00%
Ba3
16.4%
15.5%
97
89.8%
91.9%
0.02%
B3
9.0%
2.4%
6,519
54.4%
84.7%
16.65%
Baa3
54.4%
48.4%
355
84.1%
84.7%
1.49%
Ba1
23.7%
19.4%
3,174
83.2%
90.2%
5.08%
Baa2
37.3%
32.6%
146
91.7%
90.2%
0.48%
Ba1
6.6%
4.6%
2,825
20.8%
35.7%
15.32%
A2
45.8%
39.6%
247
34.5%
64.2%
1.78%
Baa1
19.8%
14.6%
Int'l SF ex COO ft Other SF
16
75.2%
79.9%
0.10%
Ba1
8.9%
4.9%
Other SF (includes SIVs)
18
61.4%
3.1%
1.71%
A2
25.8%
18.5%
US ABS ex HEL
32
85.6%
79.4%
0.20%
Ba1
6.1%
3.8%
US CMBS
14
93.4%
88.6%
0.01%
Caa1
11.6%
3.8%
509
75.4%
89.7%
2.25%
Baa3
13.6%
11.4%
46
92.3%
89.7%
0.25%
Ba1
5.2%
3.7%
209
84.2%
94.0%
0.80%
Baa2
5.8%
4.0%
excl SF COOs
Historical Averages Since 1993
US HEL (includes subprime)
excl ·05-·07 vintages
US RMBS (includes Alt-A)
excl ·05-·07 vintages
Global COOs
excl SF COOs
14
93.7%
94.0%
0.02%
Ba2
3.0%
1.2%
234
23.4%
60.9%
4.02%
A3
14.7%
10.7%
32
71.5%
69.0%
0.32%
Baa3
9.3%
5.0%
Int'l SF ex COO ft Other SF
2
79.1%
83.5%
0.01%
Ba2
4.5%
1.6%
Other SF (includes SIVs)
2
48.5%
35.8%
0.88%
Baa3
4.3%
2.4%
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~"-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~:~:::~~~~~~~~~·.·.AU9Ust 2009
':::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Introduction
In a Special Comment published in April 2003, Moody's developed a set of metrics to measure the
performance of corporate ratings with respect to the dual objectives of rating accuracy and rating stability.3
Moody's corporate rating performance report is now updated on a quarterly basis.4 Moody's first introduced
and examined its structured finance rating performance metrics in a September 2004 Special Comment
"Default & Loss Rates of Structured Finance Securities: 1993-2003," and published these performance metrics
in a stand-alone document for the first time in September 2005. The structured finance rating performance
report is now updated on a semi-annual basis.5
For both the corporate and structured finance rating performance reports, the basic unit of observation is a
monthly cohort of ratings, i.e. all outstanding ratings at the beginning of a month are recorded and their
performance tracked over different time horizons. In computing rating performance metrics for structured
finance, Moody's incorporates both the default and loss severity experience of all structured finance tranches
because Moody's structured finance ratings rank order expected loss rates. In other words, Moody's
structured rating performance metrics weigh those tranches that have become materially impaired but with
lower loss severity less than those with higher loss severity.
The most important measure of rating accuracy is the accuracy ratio, which for structured finance measures
6
the relationship between tranche ratings and their realized loss rates. This metric measures the quality of
Moody's ratings as indicators of relative expected credit loss risk.
As discussed in the April 2003 Special Comment, although relative rating accuracy is our primary objective,
Moody's recognizes that many investors are also concerned with the cardinal accuracy of the rating system.
In particular, they expect that investment-grade credits should rarely suffer credit losses and impaired
securities should normally carry low ratings well in advance of impairment. For this purpose, we regularly
track investment-grade loss rates and the average rating of securities during the 36 months prior to
impairment. Both of these measures should be low if the rating system is accurate in a cardinal sense.
We employ two measures of rating stability (or rating volatility) - the rating action rate (the frequency of rating
changes) and the large rating action rate (the frequency of rating changes of three notches or more) over a
twelve-month period. To examine how watchlist actions are used, we also report the percentages of
downgrades and upgrades preceded by watch list (review) actions in the same direction.
See ".M.~.§§.l,l.rLlJ.g.1b~..E'.§.rfQ!.!ll.§D.9.§..9.LQ.9.r.l2.Q!.§J:.§.J29.09.B§.tL09.§," Moody's Special Comment, April 2003.
4 For the latest performance report, see "I.b§..f.'.§[f9.r.m§.ol&.QLMg.Q9:l§.. gg.rP.Qr.§t.~... Q§.I;l.LB.?tlD.9~L..~.~D.§.).QQ.!t ..Q.l,l.§.r:I;§!.ly'..J,ip.g§.t§," Moody's
Special Comment, July 2009.
5 Note that the criteria used to create the data set for this report have changed from prior performance studies. The most notable changes
are that pari passu tranches are no longer collapsed and wrapped tranches are included. For a more detailed description of the data sample,
~Iease see the Appendix.
The required adjustments to convert the standard default-based AR measure to a loss-based measure are discussed in :.R.§f.?w!L~..kg.§§
Rates of Structured Finance Securities: 1993-2003," Moody's Special Comment, September 2004. The concept of the accuracy ratio is also
described in the glossary at the end of this report.
3
~~~~~~~~~:~::~~~~~~~~~·.·.AU9Ust 2009
':::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Accuracy Ratios
Figure 3 depicts the one- and five-year accuracy ratios over time with the following notable observations:
::::
For global structured finance, the one-year accuracy ratio was flat at 60.2% relative to its level six months
ago, but down from its level twelve months ago of 75.7%. Since much of the recent rise in material
impairments can be attributed to US HEL and US RMBS securities issued between 2005 and 2007, and to
SF COOs, excluding these securities as well as the Other SF sector caused the one-year accuracy ratio to
jump to 74.2%.
::::
The one-year accuracy ratios for US HEL and global COOs continued their declines. Accuracy ratios for
US RMBS and US CMBS also decreased versus their levels 6 and 12 months prior, but the ratio for US
CMBS was still close to 90% and that of US RMBS remained above 80%.
:':'
US ABS, excluding HEL, was the only major sector that exhibited stable performance over the past year,
maintaining a 96% accuracy ratio for the cohort ending December 2008. If deals that closed between
2005 and 2007 are excluded from US RMBS, its one-year accuracy ratio is also over 90%.
::::
For most sectors, the five-year accuracy ratio has not dropped as much as its one-year counterpart
because this statistic demonstrates performance on a lagged basis and the effects of the recent growth in
material impairments have not yet been fully incorporated. However, the five-year ratio also declined for
global structured finance and most of the sub-sectors.
Figure 3: One-Year (yel/ow line) and Five-Year (blue line) Accuracy Ratios
'li))';,s ,,::~~~~~Q-=:3~::~,P';-' \ t:; W-.-£ffil8S_
... _ -
l~)% ri:::--;~:G;::~.:·:;:;:::\. ,",.;:~"~:,-",,::~,,c:-
~1\f.~,
'···-·-·-··· .. :~..:.d·:""-·/_ _---..:;
(':.)%
,:~u~·~.
c\,;%
:-;C~~{:
::C~~{:
::'::?,.
+ - - - - - - - - - - - - - - - - '. .-:
+-------------+-_ _ _ _ _ _ _ _ _ _ _ __
+-------------+-------------+-------------".
'~;(
::;)
:s;
C,)
:~.:
'0
,;
'0;'
.~.
.~
:s:
m% + - - - - - - - - - - - - - txf;-:.
'>(J';~,
,:0%
3.1'%
A)%
'lO';,s
+-------------+-------------+-------------+-------------+-------------+--------------
::;.
,;
C,)
'0;'
')(x)%
9:)%
G,;
-:~;
~'~':
~
~
'"
~/).~.::.
c\,;%
:-;C~~{:
";';0
+------------+------------+-_ _ _ _ _ _ _ _ _ _ __
+------------+------------~~:~
:.-.
~
--,
f:::
~"';:
~
:.~;
,o.:J
~
~
--,
~;~;
f:::
~
S
f'.
~
~
~
,------,---_~- c----~
c.:(
__
r--::~---~~-f·':;:;;v~:::::';:;:+··,::;·=~:·:·=:;·;:~··;:-"'--~
. :.
.H
~~~~~~ ~=A i~:
e:i-:'~.
'"
~
(j
.
~-;H::.;_ r-~~> -:~-----------
'~-(?:~' ':',-'- ; . . . - - - - - - - - - - - ·::'-:f;-:• •\:;~:
.•: : : + j - - - - - - - - - - - :'O';~, +-~------------
~O'%
10%
+-------------+--------------
~~;
:'
,o.:J
~~~
~
~
--,
\~~
\~~
:0
[.,
~':
~':
.C
._)
._)
..
:~
(:;
.~~
S'
~'.~
\~~
\~~
~:~
~:~
{r":
~'.~
~:~
\~~
:.:.~:
j~:
i:.,
).j~:
Note: At the beginning of each month, securities are grouped together by their alpha-numeric ratings to form rating cohorts.
The latest one- and five-year cohorts are formed on January 1, 2008 and January 1, 2004. Breaks in the accuracy ratio
series occur when the number of impairments for the cohort is less than or equal to one. Crosses in the one-year series and
plus signs in the five-year series indicate that the accuracy ratio was computed from only one impaired security.
~~~~~~~~~~~:::~~~~~~~~~·.·.AU9Ust 2009
;:':' Special Comment :':': Moody's Credit Policy -- The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Figure 3 (continued): One-Year (yel/ow line) and Five-Year (blue line) Accuracy Ratios
u:~ ~'~:L 0;.-, '~~~i·'~~:~Vin'~:;;;::~~-;
j~;:~~ b.:::::~~,8.;L:·::;·;~::""'{;';;;;·:::::'··.•. •. :::~:7':"· . . . . •. .
c'..c::::-:::
+--------------'--/(:~t + - - - - - - - - - - - - - \:~ .• ;'0 + - - - - - - - - - - - - - ~}:.:·~·i. + - - - - - - - - - - - - - '::j'", + - - - - - - - - - - - - - :!:::'% + - - - - - - - - - - - - - +-_ _ _ _ _ _ _ _ _ _ _ __
.:~::;~~ + - - - - - - - - - - - - - {~::;~~
,):3%
OX:>S'
40%
3G%
;'0%
~O':_:)
+--------------.
+-------------+-------------+-------------+-------------+--------------
. '. . ,. . ·:·:·";;··,:;;::·:::·;···,,:·.··.;:::.~"=.£~.c,~
=-:::(~;;
30% +-------------~.• "
+-------------;~rr:_:) + - - - - - - - - - - - - - ,x}% + - - - - - - - - - - - - - ·10,%, + - - - - - - - - - - - - - :~n:) + - - - - - - - - - - - - - X.>S'. + - - - - - - - - - - - - - 10% + - - - - - - - - - - - - - !,(l'~~\
:~·:~:·~i.:.
~G(~~{
~:;;::.:·~·i.
',::),;" + - - - - - - - - - - - - - "/(:'% + - - - - - - - - - - - - - ,:;::.'.,. + - - - - - - - - - - - - - ~:~t + - - - - - - - - - - - - - ·'~~::·;0 + - - - - - - - - - - - - - 3.~~{ + - - - - - - - - - - - - - :~<.:'~·i. + - - - - - - - - - - - - - ~:j';" + - - - - - - - - - - - - - -
~"':.>s'. +-------------~,-
?O% +----~. :c-----------'.~
je>s'.
+-------------~
/,:;-'0
~::.:·~·i.
+--------': ----':-------+--------------
iiT% , - - - - - - - - - - - - - - ~:;;::.:·~·i.
(::::'%
+-------------+--------------
7:~:·~i.:. +---------~~----
+-------------,:0% + - - - - - - - - - - - - - ')),%, + - - - - - - - - - - - - - :-?(n:) + - - - - - - - - - - - - - 10% + - - - - - - - - - - - - - :~rr:_:)
8::'%
+---------~:_--____c-,
:'n';0
+----------:c---+----------:,----+----------.:-----
~~~:·~i.:. +----------:,:··c--~• c_:
::'0%
:~::·;0
0%
+-~~~~~~~~~th:~~~
.~
~~~~~~~~~:::::~~~~~~~f·AU9Ust 2009
.~
.~
.~
':':' Special Comment ":': Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
.~
.~
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Investment-Grade Loss Rates
Figure 4 shows one-year and five-year investment-grade loss rates by sector. Note that:
::::
Increases in the one-year investment-grade loss rate were seen for global structured finance, US HEL, US
RMBS, global COOs, and International SF, excluding COOs and the Other SF category. Removing the 2005
to 2007 vintages reduces the loss rate for US HEL and US RMBS by over 90%. The same is true for global
structured finance when mortgage-backed securities from these vintages, SF COOs, and the Other SF
category are excluded.
::::
In contrast, the one-year investment-grade loss rate was zero for the cohort formed in January 2008 for
US ABS ex HEL and was near zero for US CMBS.
:.:.
Because the five-year investment-grade loss rate is a lagging indicator, it still did not show the type of
increase displayed by the one-year loss rate.
Figure 4: One-Year (yellow line) and Five-Year (blue line) Investment-Grade Loss Rates
t-:.~.
t--------------:.:
;,%
t-------------~c
4?·::. t - - - - - - - - - - - - - - : ' - 2';0
t-------------c:-
.::;.
:~:
:~:
j~:
j~:
j~:
.,.
6;
-~
.::;.
j~:
;.....
."
-~
"C,:
,~',
~'-..x
-~
.~.
:.0
.',
C
(I;,
._)
c·
c
.~
:, (,% , - - - - - - - - - - - - - - -
':2:~~~
,---------------
:~.. :~..~.: . t-------}~,~./.:.:..'../.\~
":
+::;:-------------
:: C~~{: t-------:.~~~:c__----
1'
,
;.
.
~,~;. t--------,.:'"----'L.::c-:.
----
:.:':,?,.
--~
..: : - - - - " ·:":l:
t - - - - - - J . ::i'
. :.£:.i-'::···-··
.
,: ,;','. t-----.;-----~
. \.:-::- ' , - - - ....;.
Co C~~{:
....
T.,..~·T~··'·:·;-;;:~:-.....,.......,CO. '-i"-A'-,-,-~~~.... ",",'---'r'-I","
c·
g~
8; G
~;
c;
8 t
-,
.$
.$
.$
.$
.$
.~
.~
.~
c~~-:
:;'j% ..1':-.\- - - - - - - - - - - - -
:::. :)!:~~
l~
,: 4';'· :
\.
~;".':.
~.:.:-------
; ::
r._..........
'\ :.
\
:
~ .~: r':·:· ~I;.I" "';" " "' ' ' 'i; . J~:. ,.,. . ..
~;:;
::.
:'3
c';
S~
~;
,--,
.:-.-;.
.?
.?
.~
~
~
.;;
.;;
)::~",\:')(:(:'
:s:
c·
t,
.$
0
..(I;)
c.:
",",':t"I:"C) :::::'\:.0
Note: At the beginning of each month, all securities carrying an investment-grade rating are grouped together to form a
rating cohort. The latest one- and five-year cohorts are formed on January 1, 2008 and January 1, 2004.
~~~~~~~~~~;~:~~~~~~~~~·.·.AU9Ust 2009
::::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Figure 4 (continued): One-Year (yellow line) and Five-Year (blue line) Investment-Grade Loss Rates
::;% - , - - - - - - - - - - - - - - - -
A-=':~-:
;;%
~:~~#<,/
••
)••
""""/",~,
i',\";
~~::
;:."";.
:":-
m
r-..
c.'
:":-
?:
~
'
?:
(~
,~;
,~;
••
~};';
;',-;
~~:
~;.;
~~
~~:
j~:
'-"1
~'.)
(~
...
..,•.....
')
+-------------+--------------
.~
~::
...~,:()
••
••••
'::5%
••
(:,.>!'~~\
••
::.:.::;'%
••
)
t:;
~";:
(~
,~;
;'C'~',
;:~.\
m
".:.'
~~
,~;
')
(~
')
••
-,----------------
+----------------c.+--------------.-
::;::~
de,
+--------------.+--------------.c-
~
.\~
.~
,;
,;
'0;'
'0;'
C,)
C,)
-,----------------
+-------------+---------------c.
••
+ - - - - - - - - - - - - - - . -••
+--------------.+--_ _ _ _ _ _ _§i____+•
..p ......,-
"'~.'
.~
!:~~.~:; +-------------~
~2~~
~
••
,'D% ..•.. i:'-·'·····'·'·····-·-·-',···,,··~,·,',:,:,',·.'·'.'.'.•'.,..'.'.'..•.,•.•,6".'~:.:,',"".,.:':.,
~~
::',:?,. + - - - - - - - - - - - - - -:4~-;·~.
(';?%
(; '1%
;'
..........,.. '~:, . ~."'-T--'--',..,.
. .~.,..~
,.
~;::
f:
C,:
c
S S
.'.
.'.
'C'~',
.~
.~
8
::.
:'3
C)
t,
8
.~
~;
.. )
.~
-,----------------
,-,,?,. + - - - - - - - - - - - - - t·~:; + - - - - - - - - - - - - - .;-~-;.~.
Cli.
+-------------+--------------
:~.~-;.~. +-------~------,'---
4~~
+-_ _~..:.:.,:':..:..f··.~.,.:.~~:'····o••-•.••o•..:::.••.•
::?~-;.~.
C~-::{:
f-----r"r""'j""'\"{::;";":
.,
~
~~
::.
.~
~
'~(",. ..
;~.'
~;
.~
.~
'~"'....::'
-..,....:.::.~---.c-
.............,.. ,. ..,.~
:"""
.:-.-;.
C)
.~
8,.' .
5
.,
C)
'J
.~
S
(:
(:
..i\;')
..i\;')
:~:.::?O~:-:) ~-------------
.'.
\ .•~-----
:J 13%
~: ~: ': +---------cr:.~.,..:--::.:.:'.{r~s
··,··,··,··,··~·r··,··r··,··r··,··I··,··r'·:
8
8
,-r-:
..
~~~~~~~~~:;:::~~~~~~~~~·.·.AU9Ust 2009
~
~
:~~
..':\j)
ill
:~~
..':\j)
','---,----,"', ',"
~
~::
:~:
8
~",'"
e
.,
::;
:~:
.~~
~::
•••
...' ....
l
••
.1
.'
~(
\'~'
i .....
I::'
).-
i:.,
i~:
i~:
.
) -
8,
~:~
\~~
c.":
_c
. ~~
:~~:
~"':
Cv
--:'
':':' Special Comment ":': Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
\'''',
...
~;,
~"':
Cv
--:'
,.
".)
~~;
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Average Rating Before Impairment
Figure 5 presents the 36-month-average rating before impairment over time as well as the number of newly
impaired securities used to calculate the average ratings. The following observations are noteworthy:
""
The 36-month average rating before impairment for global structured finance was Baa2 for the most recent
cohort, the same average as six months prior, and one notch above Baa3 twelve months prior. Excluding
SF COOs, the Other SF category, and 2005 to 2007 vintage US HEL and US RMBS, the average rating
before impairment dropped one notch to Baa3.
":'
Global COOs and Other SF displayed the highest average rating prior to impairment of A2. For the
January 2008 cohort, the 36-month average ratings for US HEL and US RMBS were also high at Baa3
and Baa2, respectively.
,.:.
As has been the case historically, US CMBS exhibited the lowest average rating before impairment of B3
for the cohort ending December 2008. The average rating for US ABS ex HEL was the second-lowest at
Ba3.
Figure 5: 36-Month-Average Ratings before Impairment (yel/ow line) and Number of
Newly Impaired Securities (blue line)
:;.~.tKUL~~(~U~~~.~_t~~{~~l~LL~~~..lbJ~~:_~=;;:~ j G:f.';
{-\O';.,-------------...,.I.I:'~n;
U,,"J"::
~.,.,~.
;"··l·
--70;.0------:--7
.....~.-~.,..~:::,:.)
:::.,:;;::: t------------".~:::...:.< ...~...'-... ----"'----.,-,~.;
';..~:_~.'--'-;'.
·TC'
~.'
:: ...
.'.
~
:0;,
p,:.:+-----------J ::-----+
1:';,::;"0.
+-------c~--,\,".-:....:.:~1:.:.•{;·\.i",::----po
.
.
8:'.:::~ +_-------:----':.~;'""':-if
:":::3
1:';0
:~3
::
+-----~./~...: .~....,""""""'.....::..;:~.--~t:.?;"\. '-./I .
L4.'•.• OJ
().:;:1~~ ,·--y--,----"o"--,---,-----,---, -i-··,·~·-·-;·
:'-.
s;
f:::
~
~~\;
~
~"';:
f:::
~
~~}
~
\~:.-.\:-)(:~:.
,
c:
~
2;
~
::)
:'-.
~
--,
f:::
,o.:J
..)
~
~
"
}
-.-:7'
":'
(-jC<"':.
·::tK:
Gl:~;: ,.... ,."' . . . . . . . . . . . . .,. . . . . . .:. ./'''i:,:6:~:;/
,~~,
, - - - - - - - - - - - - - - - - - , :-?co
:...
'L':-':-"':;-;-;"""
.~-;,
;~~~'..::~
:1'
.:.: .. ,
.
t((!
::~;+_----------~::--+ 3C"{;(;
:~~ ~ >-::<:
.
;\~~+_c,--------------""C::t -r:'((~;
~(
.\~
~::~
C
(I;,
..
~
:~:
:s:
.
:s:
.~
cr-'
c~
.~
/":::,----,'-------------,1 r,'·
;\:i+_-------------t
·w::
~~{;:: . :i+_'--------------~:l.: -J-e;;
~::~"'",~v"\A ;
~)j
,-
..r-:)
B::
:::
:;;
.~
.';~
8
,-
..r-:)
e:::
:;;
3'·,·:1·'::1(.) O::r<:"
Note: At the beginning of each month, all securities that become impaired within the next 12-month period are grouped to
form a cohort. The latest 12-month cohort is formed on January 1, 1008. Breaks in the average rating before impairment
series occur when the number of impairments for the cohort is less than or equal to one. Crosses indicate that the average
rating was computed from only one impaired security.
~~~~~~~~~:i:;:~~~~~~~~~·.·.AU9Ust 2009
':':' Special Comment :':': Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Figure 5 (continued): 36-Month-Average Ratings before Impairment (yel/ow line) and
Number of Newly Impaired Securities (blue line)
+_--------------l-+
..../.;'.,'
",
)",
+----~,.~.'---------,,-+ ~':~~
..... "... .,.......:'/
/
. '",,/
.(
~~~
:s;
S~
.~
"-~;
:,
~
C)
::;
,'.
~:;
..(I;,
:s:
1--;-:
c
.~
:s:
+_------------}---+
./
:s:
8
,-~,
~
3~{:
~:-:
~.~;
;.:;
-,
:
-.~
-:;;
;~\:
(:
,""
")
-:.-;'
(:
~
,""
")
~1\)
E~
~:-:
1--;-:
i~.::~+_------------_+ ':;~~
[,.:,+_-------------;'-+
c~:"~~~=~i~'
icr.'
-h"""" ,..... , ''','',.'''' ...... ,",","''', ..... ''','','''' ","", ..,...... ,...;:-rl- '1
:s:
~:;;
S~
.;;
.?
::.
:'3
$;
Si
.~
")
t,
(I;
8
,'~,
t,
.$
~
(,-,k:·,t ~:;t::1r:~:lS;
::;
C)
~
(I;
")
8
~
~
~
~
8
0
8
$
~
~
\h-l-~.
L.I:.\'.:.
A','.O'-,---------~j
.'\J+_-------i,',','c--------,{+ AB;
"\·n3+-----~---c-:.
"',
,:~:;',+_-------',-."'
;.\.~~+_---~
:n:
; - - - - - - - - - - c - : ' : ' ~:..::(:
"
,,</
i "'!
.. cc----,-,c__i,'_+
l
1'~~<~,
8J+--_ _ _ _:-_'_ _-----"·•..,..-,'-ic -+ iC(j
(>-~;J':~ -r"",
·T·,·'·... (······ ...·····........... ,·
"y_',
B::
~::
.?
6:~
:,:)
S
:~:
j
:==;.,
:~.:
::;.
:s;
";'
cC·)
.'\3:·'-,--------------,1"·:
:,'-''.~....
h.~
1 .
ll;,~c'::
"
'
",: " ",D,ndil:
~~~~~~~:::~~::~::~~~~~~·.·.AU9Ust 2009
".
'~.,
:~::
j;:
j;:
i:,
j"
r.-'
K'''+_----------,-------,
;\~~+_------------_+
/
"K:
:'<:'
8::::-;+_--------------;·,''---':f :::::-:
:.':;3+--_ _ _ _ _ _ _,1_--+
<:~3O"h·.·:..
f·: ...·:..f·: ...·:..f·: ....-:..f·: ...··:..f·: ...·:..f·: ..
.-;-,
,~,
;~
l)j
'-:'
I
)~'
<~,
5~
:'?,
:'?,
:'3
~"':
~"':
---:'
._)
Cv
'--I
\~~
':':: Special Comment ":': Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
,.~;~
::::
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Rating Action Rates and Large Rating Action Rates
Figure 6 reports 12-month rating action rates and large rating action rates. Figures 7 and 8 further
disaggregate rating actions into downgrades and upgrades, and Figure 9 demonstrates how frequently
downgrades and upgrades have been preceded by watchlist actions in the same direction.7 Key observations
include:
:':'
The 12-month rating action rate in the global structured finance category was 36.2% for the January 2008
cohort, a 50% increase from the rate of 23.7% six months prior and an almost four-fold increase from 9.6%
a year ago. Most of the rating actions were large actions, i.e. changes of three notches or more, as the
large rating action rate of 31.1 % was only a little smaller than the overall rate. Excluding SF COOs, the
Other SF category, and US HEL and RMBS transactions that closed between 2005 and 2007, the general
and large rating actions rates were much lower at 13.4% and 9.6%, respectively, but still much higher than
the historical average .
.:.:
Rating changes that occurred in 2008 were comprised almost entirely of downgrades as the global
structured finance downgrade rate climbed to 35.5% while the upgrade rate dropped to 0.7%. Moreover,
the same pattern of a rising frequency of downgrades and declining frequency of upgrades was seen
across all sectors.
:,:,
The proportion of downgrades that were placed on review prior to the rating action rose to 44% for the
cohort ending December 2008, up from 35% for the cohort ending June 2008 and from 22% for the cohort
ending December 2007. The frequency of reviewed downgrades also increased over the six-month period
for all sub-sectors with the exception of US ABS ex HEL.
,:,:
In contrast, the percentage of upgrades that were placed on review prior to the rating change fell to 8% for
the January 2008 cohort from 18% for the July 2007 cohort. In addition, the rate of reviewed upgrades
was less than 10% for all sub-sectors except for global COOs.
7 Moody's also reviewed its uses of watchlist actions on structured finance securities in a Special Comment, "Structured Finance Watchlist
Resolutions: 1992-2003," June 2004.
~~~~~~~:::~~~:::~~~~~~~·.·.AU9Ust 2009
':':' Special Comment :':': Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Figure 6: 12-month Rating Action Rates (yel/ow line) and 12-month Large (three
notches or more) Rating Action Rates (blue line)
",~y;{: ~--------------
:C". t - - - - - - - - - - - - - ~~}:;{:
:::
:,,~,~;. t - - - - - - - - - - - - - - - - - - c f
:~.~;.~.::.
:~
-: :=·:~4
;;~
::,,:'0
t--------------
:e,." t - - - - - - - - - - - - - - - - - - c
..:',:,'::.
'.,. +------------~. :. :.~ :
,-
:=·:~4
c:~'
-------.'
.. .
".:
1.:::::.:::=:::. : .;: .: .:.:.: .:.: . . ::::::.:. ·"" . . . :. . . . . . .:.7:.:-:::.:-Ii';'
c(
.:-}
:,,;';'.
,.
~~\;
.:-}
G,;
.:-}
~~}
.:-}
~;~;
~;~
:--.
~'.)
~'.)
~'.)
f:::
r:;
f:::
r:;
f:::
r:;
f:::
r:;
..
~
..
~
..
~
..
....-. .'-.-.:''C'.
~--------------
..""'\:.
,;';' f:··::·,···········:···':L::::·:···:··::·····.:·.::·:·..-'
9 ". ~: ..
.:::
,~
~
~7
~---~
X;%
;;:
;;:
. . . :.:...~:.......:. . <.:~~
..,. . -;............../
'
~:
I;'
\.,
;;:
;;:
~~
~-------------
.: .....
'(;~'" t-----------~:----'-
/ .......;::
~."I
"..,",
:..:)
:..:.l
".
~"::j
".
~"::j
:J:::.:.~..
::
.:---:-
:J~t;:::::r!~~
':2;';(·
I
"
~;::
r:...
:":-
?:
~~i
).\
;'.-;
H
I
,;j
,;j
')
')
~~i
~~i
'-"1
r:...
,),]';"
~--------------
U~,% ~-------------
::-:...c;~{:
t---------------:'
:'.,;.~,;
t--------------
:::0~(':'
t--------------
..;)
. . ./;.~.::. t - - - - - - - - - - - - - - : . : -
::~;%
i'
:·: :i-:'~. t - - - - - - - - - - - - - - - - c : . ; L
::-,':?::.
(
~::C~'}-: t--------------::.~~;:
., +____________-----:. .:.l
~:.i·i;
,)';0 -, .. ;· ...., ....·.·T·.·.'·}::::····T.·bg";.....,..-;-...........,.....:.,:.,.::.:.:.:.::::.::::::.:.,.::.:)':
8
i.'
~
~
";'
~
~
";'
~
~.
q
8
8
0,
i.'
i.'
~
~
";'
•••••••••:.:.;••• ;,•• ;.;.::;.:.;,;,;;:.;.::.:.i
;:~
~::,~
"i:
"i:
0
0
,...."::.
~
8;
~
(0)
(j
~
~
();i,"-,'l :SLl'li"'(:
~~~~~~~:::~!:i:~~~~~f'AU9Ust 2009
~-:::1~::-:
.:.:. Special Comment ••:.: Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
~(~
,...
'(
c.:(
~
~
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Figure 6 (continued): 12-month Rating Action Rates (yel/ow line) and 12-month Large
(three notches or more) Rating Action Rates (blue line)
~~)~~ ~-------------
':;:~.~-;.~.
~~G'~:;
::=.:;.:)
::-.':?:,;:.
~-------------
::!
2:=·:;~
:~: -.':?: .
',%
::
;~ '
f
J
.:.~'
+----------------::-
~~: C: ·: :=:·'· ·=:'~:r:.·.;S: : : : :· · : ;:'i'~'·'·':'·' ;:, :.: .: ;. :~:t;;:;;;;t
8
~
~
~
0
2
8
~
'":~;
~~i
~~:
j~:
j~:
~~i
~~:
j~:
~~i
~
,.c;
.$
8
.$
::.::::,":.~.,~ ~~J.~=-tw~~; ~-:a~::-:
m
t,
~;
.. )
.:-.-;.
:.::)
~
..,,::)
C''-'(~~f'''f~ '::,~~~f~=)' :q
~
,.
.$
~
~
G
Cde
G~{)b::~: C[f)~-:
c:. .c~-::{:
S0~;"':' ~-------------
.:~G:;'::'
,~:}"v
+--------------
':~;~-;.:,::.
.:·.'·rv
+--------------
:~.~;~-;.:,::.
::-.':?:,;:.
::i'~:;
t;~
\'(~
e,i
~
;'C:~',
b
~"';:
~"';:
t
t
~
-:~;
~'~':
2~
~
".:J
~
'"
~
(nb':~=1
-H ;~i'~.:='-;~J
~
--,
~(~
'(
~
b
&~
~'.)
t
~::
~
:;;
-,---------------- - - - - - - - - - - -
4?·:'::.
r':' :c-~
--------------c
/
:'"
:~::-:~.----... / , / •. ,/ ··'=7.~i
(:.;:. ,C.........:... :;:...........;:. :;.:::(.. : : : :•.:)... :.". . . . . . . ./ ....:.,.........,...,. ..
tq
~:-:
\--;-:
(~
~:;
~:-:
\--;-:
9
-:.-;'
~
,""
--,
(:
:¢::
T~
T~
.~
T~
'~:
tic?
I·
'-:'
)~'
T~
:'?,
:'3
:::0~;"':'
-,----------------
~:~,.?;;
+----------------:
~:
'-";:.
I:?
;.:;
~,
~:::.
1.(
:r.
~,
.
' ' ; +-------------.:.r:
':'::'_::"v
~
;~.'
[.:~:tf·
}::::::.::.
.,
+ - - - - - - - - - - - - -.~
:::Y?v
~t--
:,. ..:;. J.........
l ••••••••• ' ( •••••••• ' ( •••••••
&~
~::
:~:
.,
;~.'
:¢::
.~~
.~~
!
:~
:L
••••••••• ' ( ••••,---,--'r••••••••
.~
.~~
tic?
'~:
'-:'
I·
'~'
.~~
:'3
:'3
Note: Rating actions include upgrades and downgrades, which are measured on the alpha-numeric (or modified) rating
scale. Rating cohorts are formed each month covering a 12-month period. The latest 12-month cohort is formed on January
1, 1008.
~~~~~~~:::~~::~::~~~~~~·.·.AU9Ust 2009
':':' Special Comment ":': Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Figure 7: 12-month Downgrade Rates (yel/ow line) and 12-month Large (three
notches or more) Downgrade Rates (blue line)
(;::oh(-:~~ ~~t=Ul..:tur~d ;:::~~'=c:~n':>:-==
",~y;{:
-,-----------------
~~~~-:.~.
+---------------:
3C::~4
.....,.
,".:-.~ I~.
<.
+-------------~::.:j
,.:=.): ::.F GO')'3; ();:n-::::' ::.F.
~~~;~/':'
·::.GiL('~L('~C'{~~"d·~~~~~iLl';;::L8aJ2-S~Jo~_SS~_
:":'0
+--------------
-
:~.:-.':?::.
+--------------{
.:.~'--
+--------------:?j
:
::'::?,. + - - - - - - - - - - - - - - : ;.;:-
:e,." + - - - - - - - - - - - - - - . :
:~·:;0
:-:.:;~
....
,;';'.
C~~{: -+: ;~....:.-.;-;-;-;:;:.:.:.:;..:.-. "i" ....:.:;.;.;:;-;-;••••:••,.,~.~,;~.•:.:•••-••;;;::.••••.•...::-::.::::::~:::;:~
g~
~;
c·
c;
.$
.$
.$ .$
C.f,,·l::):-t
:,,;';'.
8;
:.::)
r:,,::
..
~
b
r:,,::
..
~
+--------~:""".,----)-
i
:.Y~~~
~: ....
,,.-.~
:.::)
:.::)
r:,,::
r:,,::
..
~
..
c'
.::.::
~
9
.::.::
".
~:
'
;;:
;;:
;;:
..
~:
3ta:-t~:-lS-: L::.::t·~
-,-----------------
X;% - , - - - - - - - - - - - - - - - :(;~",
t--------------
:;:;~ -~·\·:·::;·:·;·----r
S~
H
::~.:
i~:
:~~:
(OJ
:..:)
:..:.l
"..,",
I
r:...
~:;;
J~
i~:
".
~"::j
".
~"::j
".
~"::j
~~i
:?,
..~;
~."I
::.
,),]';0 - , - - - - - - - - - - - - - - - - -
t,
~
Si
C)
t,
t,
8t,
..~;
..~;
..~;
~
~
~
8
.~
c·
c.;
.~
U~,%
-,----------------
+--------------
>.
':~u~·~,
+ - - - - - - - - - - - - - - - - - c.
:'.,;'~';
.4=".';'%
+--------------.{
,le,." + - - - - - - - - - - - - - -
.:f
r:':)% + - - - - - - - - - - - - - - , ; :~':..';'% +-_ _ _ _ _ _ _ _ _ _ _ _~:.;:;'--~f
f' ,,;% t - - - - - - - - - - - - - - - - ; : ..
.
.................
,.,.,
...
:i::
::i-:'~. ,.···c···,·····,··· ••;b~.iS:~·····~·T c " _ _,*,,,~~~
Ii';'
c(
~
~~~~~~~:::~~:::::~~~~~~·.·AU9Ust 2009
~;~
:.-.
~"';:
~"';:
f:::
f:::
~
-:~;
~'~':
t·?
2;
~
~
~
c:
~
(nb':~=1
-H ;~i'~.:=-;~J C.:-:tr·
~;~
~'.)
f:::
~
;:~
c:
~
3,;'~';
+--------------
",,;';(.
+-------------~..;~"
::;%
+--------------::f-/
:,Y:;(.
.....
., .. ~., .. ~~,:~,.f:···:l····:~ ....... . , .•.................................:••:•.••:;;i;
6:~
;~.'
:¢::
:;;
.';~
.';~
~::
.~
.';~
'~:
l)j
'-:'
I
'~'
.';~
:'3
:'3
::':' Special Comment :':': Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Figure 7 (continued): 12-month Downgrade Rates (yel/ow line) and 12-month Large
(three notches or more) Downgrade Rates (blue line)
o!:;::.::;~~
~
_ _ _ _ _ _ _ _ _ _ _ __
~~.0.'.
f'·'·'. + - - - - - - - - - - - - - - - - - - c
?'x,
+------------------c
"I~.
:!:::'%
2:'.'", + - - - - - - - - - - - - - - i : i
:~<.:'~·i. + - - - - - - - - - - - - - - l : : ':5%
+--------------~
I.;'••••:...;;.;.;·.;................,.... ,. ...,..."...., ••.•• , •• ,., •• ,., •• , •..,.• , ••.•.-"¥"F: ....
..
::~
:9
'.
G, ,
~
.\~
.'.
C
..(I;,
~
8!
.~
,:=:;.
.'.
~
:)
:s;
,.L
.~
:':~'};0 ~-------------
~
:s<;~
+-------------------c'
+------------------c
,::.<',;;,
+-_____________ i;:
':~% +-_____________~r
:~::·;0 +----------------i~
~~;: ,
_ _ _ _ _ _ _ _ _ _ _ __
+--------------
·,':I.:
,'.',·.
tl:'~~"'~
,A ' .
""l
·····::·::,··:·::r···;::·:···~···.. ·,···:·;···'·····r...,.....;.~•.•••'.'.'.:.•:~.~.'.';:•••:••'
~~
~~;
2;
~;~
~
~
~
0
"i:
..
C':
".:J
::)
".:J
23
0
~
---,
~
---,
~
::-:
~%~------------~~~;-=:t':'
+-_ _ _ _ _ _ _ _ _ _ _ __
:'n';0 + - - - - - - - - - - - - - - < , ; -
::::;% + - - - - - - - - - - - - - -
:,<.:".,. +-_ _ _ _ _ _
:«;% +------------c:i~···· . : - - - - - -
~.~c_---_::'-r
.:........... .
.if
"i'-'----'''...,.,.----':-
:::.:".,. + - - - - -
.,' :"
.;.::.: {~""
..~;~:.:~.~:..:.:'.:,.-/
-'"
,-
~::.~
"i:
0
-:?';0
\<:.
~~
0
~
"i:
\....
..•.:.::;.. ................:..,:. .:i:.
.
(:j
~:~;
(.>
f
0
~
~
",
......,..
:
.""~, +-----.~/\.\.:
.:......... . . . . . . . . . .:. ,. .l
::.::).
(!;.;;~
~,
,~
~
(;:;0
.::
.
....
.-~--------------;
:;. J. =!:,·.
8
~
~
~
: :r.:,:.i:8,.: ; ;>· : : · ~:;:<:.: .:. . .. .y
~
::.::.
~
~
~
~
_
8
S
.;;.;;::.
~
~
~
~:
.~
S?
S?
...~~)
...~~)
.l
0:::
.,.
'~ ..
:~::
.::;.
J;~
J;~
J;~
~:~ ~:;
: .:'
:«;%
·'::~t
.;/::....
··,··I··,··I··,··I··,··,··,·~~"--r,~~~W";=~
0:::
~-------------
",
+----------------c
~----------r
',::: + . . '.
.i"
;~.~ j .•.•. ;..... ,.•. . ,.::: .• ,.. ,:;, . . ,.:.:;::. . /.::;,.,., ..~.. ,....,L
.")
~J)
~S
. .
0:::
."
---,
0:::
."
---,
"
~:;
R~
~:~~
I?:
I?:
I?:
Note: Downgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month
covering a 12-month period. The latest 12-month cohort is formed on January 1, 1008.
•~.~.~.:::.~.:~::~.~.f·.AU9Ust 2009
'II::
Special Comment :':': Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Figure 8: 12-month Upgrade Rates (yel/ow line) and 12-month Large (three notches or
more) Upgrade Rates (blue line)
~:.:;~~ - , - - - - - - - - - - - - - - - -
X : 0 / - . - - - - - - -..c-.- - -
S~
.;;
.?
~: ....
,;;-;.
r:,,::
..
~
;~.'
"~
8;
,$
:.::)
r:,,::
..
~
.:-.-;.
~
8
,$
,,.-.~
c'
C)
.:::
t,
~;
,~
")
~o
m
0:::
.~~
")
~:.:;~~ - , - - - - - - - - - - - - - - - -
~~n'0
,"
~:
:s:
J;~
;~:
C)
0:::
~~
.~~
")
-,----------------
+---------------c';.. . . +--------------c.i ....
~?% +----c----------c.'.... - ' , ~(-)':~0
.en:; + - - - - - - - - - - - - - -
".,".;.
+-:-,~,---------~-:,:-
~
is:
0:::
.~~
"')
'
~o,%
'7.:'
1;-
".
:~i::
:\:
G;
:.~;
r:...
~;
H
:'~.'
~3
~3
-.S:
i~:
i~:
j~:
j~:
~~i
~~i
~~i
")
S
;:~
\.,
-,
-,
r.,··!",c,·:t ~:;t::1r:~:lS;
;)" I·n.
;.lSl·n.
,'.'",
-,----------------
~.
,v
. 1.·~i':.
.
+--------------
-.~~-:
·~~~t
+--------------
~~'}-:
·~:~t
+--------------
2:~~~
i
"'"
,;j
\h-l-~.
).,1:.\',:.
'0:')·'(:',:- V~~~~·~g.:=~s.
;:;(.
:;!:~~
c"
C";
.~
::;)
....
.--:.
-~
'0
r:C,)
;:~
"i:
0
~~:~
:'-.
~
~
"')
~"'; ....
f:::
~B
~
"')
~'.)
f:::
~
::.;
c:
~
~):-l'::..}:i ~:;t8!:i~~~~_; =)-~=.~
~~~~~~~:::~~::~:;~~~~~~·.·.AU9Ust 2009
':':' Special Comment :':': Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
~
c:
~
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Figure 8 (continued): 12-month Upgrade Rates (yel/ow line) and 12-month Large
(three notches or more) Upgrade Rates (blue line)
?~~~ ~-------------
I,,;,
~--------------
+----c----------\"':
,'.';" +-------c.::-:--.::c-.- - - - - - - - c : ; - - - -
')%
~;·"v
+--------------.::.:'".
.
+--_'-:-- .:---- - - -
'.::;;;
,"
f'%
::,%
--~:.<---
:".
:~:'
::bt~~
:s:
~:
):;.
~:
;;:
~~
;;:
.;'
\.,
~
~~
;;:
~
~~JjbA~
•
\
.'
T
~.:
;:."";.
~
~~:
~-4-::{: ~-------------
~2~------------} i
-------------
..·r:~
~•.:::. ..,.... ,.... ,.... ,::::}'#.r.:·:·:,·:·::::,,·:·,·:,,·:;.:::.::.:.......:;:.:. ~:=
~;
)"'<.
:.;:
:.;:
'~ ..
~:
,"
)~'
;~
;~::
.;~~
:'?,
:'?,
T~
---------~.L-
~: : : ,... :, .. ,.,..:,.,..i:;:;LG.,.:~"".:",,·:·::::::j·:. . .:;:.",.,\,:.:.
---------------
_________ G __
'-'I
.~~
._)
:~::
)~'
:~:
ti...~
'-;'
.,
::;
;;:
:'3
;;:
:'3
.~~
c~
.C
~':
\~~
'l%
+ - - - - - - - - , . .. /.-.- - - - - -
')%
+-------::'--::-~----
2%
:~
,,'
';% - , - - - - - - - - - - - - - - - - -
'r;,; + - " : ' : , - - - - - - - - - - - - -
~\'
.C
.~~
t··
S% - , - - - - - - - - - - - - - - - -
.~
(:":
+-----:-\-~~ :->".-----
':",",LL/J~!Jt,LtX'
i
::.~
~
i.'
~
M
";'
8
i.'
~
M
";'
8
i.'
$
~.
$
~
Note: Upgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month
covering a 12-month period. The latest 12-month cohort is formed on January 1, 1008.
~~~~~~~:::~~~:l~~~~~f·.AUgust 2009
::':' Special Comment :':': Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Figure 9: Percentages of Downgrades (yel/ow line) and Upgrades (blue line) Preceded
by Watchlist Actions in the Same Direction
'IU";;, - , - - - - - - - - - - - - - - - -
:\
............. .
;':;
j~+--r,.,-.;.,.-,)-;--:-;---;ft:~
~C<;-=:t0
~; l~)t~
i
Si
~;
$;
~f.
~;
")
.~
.~
~
,"'\
-,'\
:'-.
(~
S'(~
.~;
.~;
t,
:-.:~
<:~
,;j
,;j
')
')
~";:
t,
8t,
~;
~;
")
")
C)
8
,$
"Rl::.Q.-.C8f~.~6j;::~~i.-tl~t:B.---'-~~~t8.~
l
,';'
:::;,
+ .. '"•••
->'-r-,~~~~~~~~~~~
."
...~~)
J;~
:s:
...~~)
,;;-;.
~: ....
S~
S~~
,'~,
"~
:?,
0:::
"~
c.::
~:
B
(:3
(;
:......
'.'
':.:.,
~~
j~:
j~:
j~:
i~:
"~
r:,,::
";'
.~
ffi
0:::
,~
:
::.
,~'.
,,'
:s:
8
.~
G~:~i"\~~ :~.t:l=---::-W~q :.»~0
L~;
:'E,
·l(~·};,f.
~ C(;-=:~~
t~~:·~t':.
E::f:~~
,'.'-.'"J~
1.1'.'i(_
,"::'%
"K}~t
~K:~:-;;
··.··.·:u
.-= •..•.. (.
:':C~;;.f~
n%
s;,
~
~\;,
~
,...."::.
~
8;
~
(.>
~
C:.--:·:)or!: ~-=;!:3:i~~~~.:
~~~~~~~:::~!;~::~~~~f·.AU9Ust 2009
~~
'(
".c.:(
~
~
~
~(~
Ci;.f~
:Ui~~
t1
. ,. .=~,;C[.~:.m,. ,. . . . . . ,. ~~~~~
'.-t
S?
:~::
..':\j)
j;:
:~:
'
..
~\'
5~
,~.
:~~
..':\j)
~::
:;;
~: ....
::-;.
:,:)
5~
..r>:)
,~
:-j~=.-~
::::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
t:
.;~~
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Figure 9 (continued): Percentages of Downgrades (yel/ow line) and Upgrades (blue
line) Preceded by Watchlist Actions in the Same Direction
~,~3%
~~:
~(
"G,:
"C,:
.',
..,
8
.\~
.\~
.\~
.\~
(I;,
(I;,
c
..
~
:.0
.',
c
.',
~j~
c
C
;:."";.
~
~
~
..
(I;,
..
..
-,----------------
~: : ,:
n
t-----cC--
7rJ
~
·~\O%, t----,~-.r;_-·-·'+!-- - _ J _.. ' - - - - - - - - ~ ~1 i~
Si
;:~'
,..J
~
.$
(.\S~, --r·l··,··l··r"·r··c··r··r"·;:---,----r:·i··'··:··;'~·~~~~~~
t,
(I;
.. )
G
.\:
~
;-< •
~;~;
(~::
;.":'
:":-
".
~"::j
?:
(~
i';
;:."";.
~
((::
:":-
?:
=~~~r:~0
-,----------------
('0%
t-------; :C-.~---;.c__--~~,-,,2~::\.;.
.-:
,~)% t - - - - - , c---c"-'---;c-co'·~····~·······.7'·'~
UP
~
(I;,
.::
.. ..:;;;':
.:,./~ :
~:; :.f~ ·, +----.-.~.~i/llt:(L~t
~~(r:~0 +-------:.•'--:•.::J~:.
:!
/0%
0%
.~
.~
5~
:~':
is:
:;.:....
.::··/~.1
t.,.::. ;..:~·-··~/~~~~~~
""'::""::""""TJ
c(
t
.
~ ~-----
&:
m
.::
0:::
.~~
0
"")
:,).;~~~-~~-:
'~.'
is:
3b:fI:':,)
C;t(:0r
S
(:
i\)
"")
8
.::
.::,
0
-.S:
8
.::
.::,
Q'.::.::~
~5:
=~~~r:~0
t(l'~~\
~:;;:.f~~·,
..~O'%
-~\O%,
t----c;-:;c-:.-+---/'-'•: f f i - - - - " 7 : -
.:::OS~I
0%
;:~'
~
Si
t,
~;
")
G
,$
,~
c(
5~
:~':
is:
m
(:
.~~
"")
:,).;~~~-~~-:
&:
.::
0
'~.'
is:
3b:fI:':,)
S
(:
i\)
"")
0
-.S:
Q'.::-::~
Note: At the beginning of each month, all securities that experienced downgrades (upgrades) within the next 12-month
period are grouped together to form a rating cohort. The latest 12-month cohort is formed on January 1, 1008. Gaps in the
data indicate that there were no downgrades (upgrades) during that time period.
~~~~~~~:::~~:i;::~~~~~~·.·AU9Ust 2009
':::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Appendix: Description of Data Sample and Glossary
The data sample used in this report includes all public, 144A, and private tranches with a published Moody's
long-term global debt rating among global asset-backed securities (ABS), commercial and residential
mortgage-backed securities (CMBS and RMBS), collateralized debt obligations (COOs), and other structured
finance, including asset backed commercial paper (ABCP), structured investment vehicles (SIVs), structured
covered bonds, catastrophe bonds, and derivative product companies. Provisional ratings, credit estimates or
evaluations, short-term ratings, and national scale ratings are not included. The following types of securities
are excluded from the definition of global structured finance and therefore are not included in the data sample:
repackaged securities, structured notes, and other credit derivatives which are basically pass-throughs of the
rating of another entity.
This data set is an expansion of the data set that was used in prior structured finance performance studies. 8
In particular, this data sample:
::
Includes tranches wrapped by financial guarantors, government agencies, and government sponsored
enterprises (GSEs);
::
Includes interest-only (10) and residual tranches;
:.:.
Includes some transactions outside of the four major sectors (ABS, COO, CMBS, RMBS) of structured finance,
such as ABCP, SIVs, structured covered bonds, catastrophe bonds and derivative product companies;
:':'
Does not collapse tranches with the same rating from the same deal, i.e. all pari passu tranches are
counted in the data sample. The exceptions to this are notes with the same rating issued out of the same
program for ABCP, SIVs and structured covered bonds, in which case only the rating of the program and
not each individual security is counted.
The data used to create this report are commercially available via Moody's Structured Finance Default Risk service.
For more information, please email DefaultResearch@moodys.com.
.~.~.~~~.~.~y.........................................................................................................................................................................................................
Payment Shortfall
Structured finance securities are defined as having a payment shortfall (previously called "payment default") if
they have experienced either one of the following:
::
Interest shortfall, or
::
Principal write-down/loss.
Reductions in interest paid that arise due to prepayments of principal on the underlying loans or due to
limitations imposed by "available funds caps" (AFC) are not considered to be interest shortfalls. On the other
hand, "payment-in-kind" (PIK) events, in which the interest payment is deferred and capitalized into the
balance, are treated as interest shortfalls, regardless of whether or not it is described as a default event in the
bond's indenture. Explicit principal write-downs are included whereas implicit principal write-downs or undercollateralizations are not.
Material Impairment
Structured finance securities are defined as being in material impairment if they have:
8
:.:.
Sustained a payment shortfall that has not been cured, or
.:.:
Been downgraded to Ca or C, and hence is expected to suffer a significant level of payment losses in the
future.
The expanded data sample was first introduced in our 2007 rating transitions studies.
I1IlrrAugust 2009 ':':' Special Comment :':': Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
'::::::::::::::::::::::: ...
The Performance of Structured Finance Ratings: Full-Year 2008 Report
The impairment status of a security may change as it goes from cured (i.e. all outstanding shortfalls and losses
were repaid in full) to uncured (i.e. positive interest shortfalls or principal losses outstanding), or vice versa. If
a security downgraded to Ca or C, but not in payment shortfall, is subsequently upgraded, then it is no longer
in material impairment. Securities downgraded to Ca or C that are not upgraded are in material impairment
even if their payment shortfalls have been cured. Finally, securities with very minor shortfalls or losses are
excluded.
Principal Impairment
This refers to materially impaired securities that have experienced principal write-downs or principal losses, or
have been downgraded to Ca or C even if a principal write-down or loss has not yet been observed. In
particular, if a security has experienced principal write-down/loss or was downgraded to Ca or C, it is called a
principal impairment regardless of whether it has experienced interest shortfalls.
Interest Impairment
This refers to materially impaired securities that have experienced only interest shortfalls, no principal losses,
and were not downgraded to Ca or C.
Investment-Grade (IG) and Speculative-Grade (SG) Ratings
Investment-grade ratings refer to Aaa, Aa1, Aa2, Aa3, A1, A2, A3, Baa1, Baa2, and Baa3. Below investmentgrade or speculative-grade ratings refer to Ba1, Ba2, Ba3, B1, B2, B3, Caa1, Caa2, Caa3, Ca, and C.
Loss Severity or Loss-Given-Default (LGD)
The LGD rate of an impaired structured finance security is measured as the sum of the present values of net
losses, including both interest shortfalls and principal losses, discounted by the security's coupon rate and
expressed as a percentage of a given principal balance such as the principal balance at origination, at the
impairment date, or at any given cohort date.
Accuracy Ratio (AR)
An accuracy ratio based on the loss experience of structured finance securities (or an accuracy ratio adjusted
for loss-given-default, or LGD) is the ratio of the area between a loss-based cumulative accuracy profile (CAP)
curve and the 45-degree line to the maximum possible area above the 45-degree line. The loss-based CAP
curve (or a CAP curve adjusted for LGD) plots, for each rating category, the proportion of the losses of all
impaired securities accounted for by securities with the same or lower rating against the proportion of all
securities in the sample population with the same or lower rating. 9
To calculate accuracy ratios, rating cohorts are formed for each calendar month over the study period so that
all outstanding securities in the sample population are grouped together by their alpha-numeric ratings at the
beginning of the month. For each monthly rating cohort, we determine the number of securities that became
impaired within one or five years of the cohort formation date and their loss severity rates as a percentage of
the cohort date balance. Cumulative shares of securities rank-ordered by rating are calculated for the
universe of all securities and the universe of impaired securities, respectively, and based on this, a CAP curve
is plotted. Note that only LGD for principal impaired securities are used in the calculation. Please see the
definition of LGD for further details.
The CAP curve adjusted for LGD is also known as a "power curve" because it shows how effective a rating
system is at differentiating between securities that have sustained high losses from securities that have
sustained low or no losses. The metric is defined relative to the distribution of ratings in the population. The
accuracy ratio is an effective way to summarize the CAP curve into a single number. The accuracy ratio is
zero if the CAP curve collapses to the 45-degree line, suggesting that all impaired securities are randomly
distributed throughout the population without regard to rating.
9
For an illustration of the CAP curve adjusted for LGD, see "Default & Loss Rates of Structured Finance Securities: 1993-2003," Moody's
Special Comment, September 2004.
~~~~~!:i:~~~:::~~~~~f·.AU9Ust 2009
':::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Investment-Grade Loss Rate
The one-year investment-grade loss rate is calculated as follows. First, for a given cohort, we compute the
LGD as a share of the tranche balance as of the cohort date for each security that carried an investment-grade
rating at the cohort formation date and became impaired within a 12-month period after the cohort date. We
then take the sum of these LGD rates and divide by the total number of investment-grade securities
outstanding as of the cohort formation date. Note that the LGD rate is not weighted by dollar volume and that
only LGD for principal impaired securities are used in the calculation. The five-year investment-grade loss rate
is calculated similarly.
Average Rating Before Impairment
The rating of an impaired security is measured every month for 36 months prior to impairment. These 36
rating measurements are averaged together to create one representative number for each impaired security.
For a particular cohort, the average rating before impairment is the weighted average of these average ratings
for each security that became impaired within 12 months after the cohort formation date. The weight for each
security is the LGD rate of the tranche as a share of its original balance. This weighting scheme will place
greater emphasis on the average ratings of impaired tranches with higher LGD over impaired tranches with
lower LGD. 10 Note that only LGD for principal impaired securities are used in the calculation.
Rating Action Rate (Downgrade and Upgrade Rate)
The rating action rate is defined as the number of securities that experienced a rating change within a year
after cohort formation divided by the total number of securities outstanding at the cohort formation date (the
beginning of each month). Rating changes are measured on the alpha-numeric rating scale and are based on
comparing the rating at the beginning and end of the time period under consideration. However, if a rating
was withdrawn by the end of the time period, then the rating prior to withdrawal is used as the end rating.
Note that a security will only be counted if it was outstanding as of the cohort formation date. Downgrade
rates and upgrade rates are measured similarly based on downgrade and upgrade rating actions, respectively.
Large Rating Action Rate
A large ration action is said to occur if a rating action (or cumulative rating actions) cause(s) a security's rating
to change by three or more notches within a year after cohort formation. The large rating action rate is the
number of such securities divided by the total number of securities outstanding at the cohort formation date.
Large downgrade rates and large upgrade rates are measured similarly based on large downgrade and large
upgrade rating actions, respectively.
Percentage of Downgrades (Upgrades) Preceded by Watchlist Actions
in the Same Direction
This metric is defined as the total number of downgraded (upgraded) securities that were placed on the
watch list in the same direction before they were downgraded (upgraded), divided by the total number of
securities that were downgraded (upgraded) within 12 months after the cohort formation date.
ABS ex HEL
ABS stands for asset-backed securities. This structured finance sector includes securities backed by both
traditional asset types such as auto loans, credit card receivables, student loans, and manufactured housing
loans, and non-traditional asset types such as mutual fund fees, tax liens, tobacco settlement payments, and
intellectual property. Home equity loans (HEL) are explicitly excluded from US ABS ex HEL.
10 We began using LGD rates as weights in computing the average rating before impairment in the full-year 2005 structured finance
performance report. Ideally, LGD rates should be calculated as a percentage of the principal balance outstanding for each month in the 36
months prior to the impairment, and ratings should then be weighted by these monthly LGD rates. Practically, however, it does not make a
substantial difference in the average rating number whether we use LGD rates as a share of impairment-date balance, original balance, or
monthly principal outstanding. Since the monthly LGD rates are very time consuming to compute due to amortization, we use the LGD rate
as a share of original balance as the weight variable.
tllJ~:~rAugust 2009::::
'::::::::::::::::::::::: ...
Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
HEL
The home equity loan or HEL sector includes securities backed by subprime (B&C) mortgage loans, home
improvement loans, high loan-to-value (high LTV) loans, home equity lines of credit (HELOCs), and closedend second-lien loans, as well as net interest margin (N 1M) securitizations. It does not include securities
backed by Alt-A mortgages, which are included in the RMBS sector. HEL is part of the ABS sector.
Prior to 1998, RMBS collateral was generally defined as first-lien residential mortgages, regardless of the
credit quality of the borrower. HEL collateral generally included junior liens such as HELOCs or closed-end
seconds. However, as subprime lending became more prevalent, the market shifted its definition such that
HEL encompassed subprime first-lien residential mortgages while RMBS included first-lien mortgages made to
higher quality borrowers. Since 1998, a deal classified as RMBS by Moody's is generally backed by prime or
Alt-A quality first-lien residential mortgages, while a deal classified as HEL is generally backed by subprime
first-lien mortgages or junior liens. Therefore, a subprime deal which would be classified as HEL today may
have been classified as RMBS in the past.
RMBS
RMBS stands for residential mortgage-backed securities. The vast majority of these securities are backed by
first-lien prime mortgages or by Alt-A mortgages. For further details, see the definition of HEL.
CMBS
CMBS stands for commercial mortgage-backed securities. Commercial real estate (CRE) COOs, where 70%
or more of the collateral is comprised of CRE loans, are classified as CM BS. If the collateral backing the
transaction contains less than 70% CRE loans, then the deal is classified as a COO.
COOs
COOs stand for collateralized debt obligations. Derivative securities such as structured notes and repackaged
securities are not considered to be part of this sector. Commercial real estate (CRE) COOs are also excluded
(see the definition of CMBS).
Other Structured Finance
Other structured finance consists of structured finance securities not categorized in the five major sectors
(ABS ex HEL, HEL, RMBS, CMBS, and COO) including asset-backed commercial paper (ABCP) programs,
structured investment vehicles (SIVs), structured covered bonds, insurance-linked securities such as
catastrophe bonds, and derivative product companies. However, notes carrying only short-term ratings such
as commercial paper are excluded.
Global Structured Finance
Global structured finance captures securities issued around the world in the five major sectors - ABS ex HEL,
HEL, RMBS, CMBS, and COO - and in the Other Structured Finance category.
US Structured Finance
US structured finance securities are denominated in US dollars and issued in the US market or denominated
in Canadian dollars and issued in Canada. In cases where the source of the underlying collateral and the
denomination of the securities cross multiple countries/regions, deals are classified by the location at which
they are monitored.
Inti SF ex COO and Other SF
This refers to securities that are not denominated in US dollars and issued in the US market and not
denominated in Canadian dollars and issued in Canada. The majority of the securities in this sector are
issued in Europe, the Middle East, and Africa (EMEA); the rest are issued in the Asia Pacific region and Latin
America. COOs and Other SF are excluded.
tlttt~rAugust 2009::::
'::::::::::::::::::::::: ...
Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Moody's Related Research
Special Comments:
,:,:
The Performance of Structured Finance Ratings: Mid-Year 2008 Report, November 2008 (112347)
:,:,
The Performance of Structured Finance Ratings: Full-Year 2007 Report, July 2008 (110179)
,:,:
The Performance of Structured Finance Ratings: Mid-Year 2007 Report, October 2007 (105390)
:,:,
The Performance of Structured Finance Ratings: Full-Year 2006 Report, May 2007 (103017)
':':
The Performance of Structured Finance Ratings: Mid-Year 2006 Report, September 2006 (99034)
:':'
The Performance of Structured Finance Ratings: Full-Year 2005 Report, May 2006 (97346)
':':
The Performance of Structured Finance Ratings: Mid-Year 2005 Report, September 2005 (94463)
:':'
Default & Loss Rates of Structured Finance Securities: 1993-2008, August 2009 (119617)
':':
Measuring Loss-Given-Default for Structured Finance Securities: An Update, December 2006 (101284)
:':'
Structured Finance Rating Transitions: 1983-2008, March 2009 (115157)
':':
Japanese Structured Finance Rating Transitions: 1994-2008, March 2009 (115070)
:':'
Asia-Pacific (ex-Japan) Structured Finance Rating Transitions: 1990-2008, March 2009 (115165)
':':
EMEA Structured Finance Rating Transitions: 1988-2008, April 2009 (116507)
:':'
Deal Sponsor and Credit Risk of U,S, ABS and MBS Securities, December 2006 (100872)
':':
The Relationship between Par Coupon Spreads and Credit Ratings in US Structured Finance, December
2005 (95494)
':::
Structured Finance Watchlist Resolutions: 1992-2003, June 2004 (87305)
,:,:
The Performance of Moody's Corporate Debt Ratings: June 2009 Quarterly Update, July 2009 (118748)
:,:,
Measuring the Performance of Corporate Bond Ratings, April 2003 (77916)
,:,:
Guide to Moody's Default Research: June 2009 Update, June 2009 (118044)
To access any of these reports, click on the entry above, Note that these references are current as of the date of publication
of this report and that more recent reports may be available, All research may not be available to all clients,
Ilt::::rrAugust 2009 ':::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
'::::::::::::::::::::::: ...
The Performance of Structured Finance Ratings: Full-Year 2008 Report
Report Number: 119780
Julia Tung
Geraldine Kim
Alisa Llorens
© Copyright 2009, Moody's Investors Service, Inc. and/or its licensors and affiliates including Moody's Assurance Company, Inc. (together, "MOODY'S"). All rights
reserved. ALL INFORMATION CONTAINED HEREIN IS PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR
OTHERWISE REPRODUCED, REPACKAGED, FURTHER TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED
FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN WHOLE OR IN PART, IN ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER, BY ANY
PERSON WITHOUT MOODY'S PRIOR WRITTEN CONSENT. All information contained herein is obtained by MOODY'S from sources believed by it to be accurate
and reliable. Because of the possibility of human or mechanical error as well as other factors, however, such information is provided "as is" without warranty of any
kind and MOODY'S, in particular, makes no representation or warranty, express or implied, as to the accuracy, timeliness, completeness, merchantability or fitness
for any particular purpose of any such information. Under no circumstances shall MOODY'S have any liability to any person or entity for (a) any loss or damage in
whole or in part caused by, resulting from, or relating to, any error (negligent or otherwise) or other circumstance or contingency within or outside the control of
MOODY'S or any of its directors, officers, employees or agents in connection with the procurement, collection, compilation, analysis, interpretation, communication,
publication or delivery of any such information, or (b) any direct, indirect, special, consequential, compensatory or incidental damages whatsoever (including without
limitation, lost profits), even if MOODY'S is advised in advance of the possibility of such damages, resulting from the use of or inability to use, any such information.
The credit ratings and financial reporting analysis observations, if any, constituting part of the information contained herein are, and must be construed solely as,
statements of opinion and not statements of fact or recommendations to purchase, sell or hold any securities. NO WARRANTY, EXPRESS OR IMPLIED, AS TO
THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR
OTHER OPINION OR INFORMATION IS GIVEN OR MADE BY MOODY'S IN ANY FORM OR MANNER WHATSOEVER. Each rating or other opinion must be
weighed solely as one factor in any investment decision made by or on behalf of any user of the information contained herein, and each such user must accordingly
make its own study and evaluation of each security and of each issuer and guarantor of, and each provider of credit support for, each security that it may consider
purchasing, holding or selling.
MOODY'S hereby discloses that most issuers of debt securities (including corporate and municipal bonds, debentures, notes and commercial paper) and
preferred stock rated by MOODY'S have, prior to assignment of any rating, agreed to pay to MOODY'S for appraisal and rating services rendered by it fees ranging
from $1,500 to approximately $2,400,000. Moody's Corporation (MCa) and its wholly-owned credit rating agency subsidiary, Moody's Investors Service (MIS), also
maintain policies and procedures to address the independence of MIS's ratings and rating processes. Information regarding certain affiliations that may exist
between directors of MCa and rated entities, and between entities who hold ratings from MIS and have also publicly reported to the SEC an ownership interest in
MCa of more than 5%, is posted annually on Moody's website at www.moodys.com under the heading "Shareholder Relations - Corporate Governance - Director
and Shareholder Affiliation Policy."
Moody's Investors Service
~~~~~!:i:~~~:~::~~~~f'AU9Ust 2009
':::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Full-Year 2008 Report
Special Comment
December 2009
Table of Contents:
Highlights
Introduction
4
Accuracy Ratios
5
Investment-Grade Loss Rates
7
Average Rating Before Impairment
9
Rating Action Rates and Large Rating
Action Rates
Appendix: Description of Data Sample
and Glossary
Moody's Related Research
The Performance of
Structured Finance Ratings:
Mid-Year 2009 Report
11
Highlights
20
24
This Special Comment updates Moody's structured finance rating performance
metrics as of June 2009. The highlights of this report are:
:':
Overall, 12,026 structured finance securities became impaired in the
first half of 2009: 106 in US ABS, excluding HEL, 377 in US CMBS,
3,218 in US HEL, 6,295 in US RMBS, 2,004 in global COOs, 22 in the
international structured finance sector excluding COOs and SIVs and
Other SF, and 4 in SIVs and Other SF. Of these, 11,548 were principal
impairments (experienced principal losses or were downgraded to Ca or
C), while the remaining 478 were interest impairments (experienced
interest shortfalls only).
:':
For global structured finance, the one-year accuracy ratio declined to
57.4% from 59.7% six months ago (see Figure 1). Excluding the most
troubled sectors - structured finance COOs (SF COOs), the SIVs and
Other SF category, and US HEL and RMBS securitized between 2005
and 2007 - caused the one-year accuracy ratio to increase by 5.4
percentage points to 62.8%.
:::
The five-year accuracy ratio also decreased from 73.8% for the cohort
Analyst Contacts:
New York
1.212.553.1653
Julia Tung
Vice President - Senior Credit Officer
Nicolas Weill
Group Managing Director - Chief Credit Officer
ending December 2008 to 59.0% for the most recent cohort.
:.:
The one-year investment-grade loss rate increased to 9.9% for the
cohort ending June 2009, a 20% increase from its six months-prior rate.
Moody's Investors Service
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
:,:,
The average rating during the three years prior to impairment for all impairments occurring between July
2008 and June 2009 was Baa1. This was one notch above the average rating for the cohort ending six
months earlier and twelve months earlier.
':':
The one-year rating action rate rose to 58.3%, a 61 % increase from the rate of 36.2% six months prior and
2.5 times larger than the rate of 23.6% a year ago. The large rating action rate was 49.9% indicating that
most of the rating actions involved movements of three notches or more. The increases in both rates were
caused by growth in the number of downgrades, which affected all sectors of structured finance.
:':'
Almost all sectors of structured finance experienced declines in their accuracy ratios and increases in their
one-year investment-grade loss rates (Figure 2). However, US ABS, excluding HEL, and US CMBS
maintained one-year accuracy ratios of above 85%, and their one-year investment-grade loss rates for the
most recent cohort were low at 0.24% and 0.04% respectively.
Figure 1: Summary of Global Structured Finance Rating Performance as of June 2009 1
Number of
New Impairments over
Prior 12
Months
1-Year
Accuracy
Ratio
June 2009
20,118
Cohort Ending Date
5-Year
Accuracy
Ratio
1-Year
InvestmentGrade Loss
Rate
36-Month
Average
Rating Before
Impairment
1-Year
Rating
Action
Rate
1-Year
Large
Rating
Action
Rate
57.4%
59.0%
9.89%
Baa1
58.3%
49.9%
Global Structured Finance
December 2008
12,738
59.7%
73.8%
8.27%
Baa2
36.2%
31.1%
June 2008
6,644
60.0%
77.3%
5.05%
Baa2
23.6%
20.6%
December 2007
2,146
75.3%
79.6%
1.09%
Baa3
9.6%
6.4%
Average (1993·Most Recent)
1,704
63.1%
80.1%
2.28%
Baa1
12.8%
9.7%
Global Structured Finance excl SF COOs, SIV and Other SF, and '05-'07 vintage US HEL 8: RMBS
2,820
6Z.8%
74.0%
Z.46%
BaaZ
33.9%
26.8%
December 2008
888
72.3%
83.0%
0.68%
Baa3
13.4%
9.6%
June 2008
424
82.4%
81.Z%
0.26%
Ba1
10.0%
6.9%
December 2007
219
87.0%
81.0%
0.11%
BaZ
5.9%
2.6%
Average (1993·Most Recent)
270
80.1%
82.7%
0.24%
Baa3
6.8%
3.9%
June 2009
:..
~..............................................................................................................................................................................................................................................................................................................................................................................................:........................................................................................................................................................................................................................................................................................................................................................................................................................................................................................ ....................................................................................................................................................................................
1 A glossary appears at the end of this report. The number of impairments for historical cohorts is subject to revision during each update as
payment shortfalls can be cured and past remittance or trustee reports may be revised. In addition, consistent with Moody's annual default
and loss study, Moody's now derives loss rates using loss-given-default (LGD) from principal impaired securities alone. The historical
average of the number of new impairments over the prior 12 months is calculated as the total number of newly impaired tranches divided by
the number of years in the sample period, and has been rounded to the nearest integer unless rounding results in zero.
~~~~~~~~~:::;:~~~~~~~~~·.·.December 2009
:':': Special Comment ':':' Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Figure 2: Summary of Structured Finance Rating Performance by Sector as of June 2009
Cohorts Ending June 2009
US ABS ex HEL
US CMBS
US HEL (includes subprime)
excl '05-'07 vintages
US RMBS (includes Alt-A)
excl '05-'07 vintages
Global COOs
excl SF COOs
Int'l SF ex COO
a Other SF
124
86.4%
84.9%
0.24%
Ba3
26.2%
17.9%
426
87.6%
92.8%
0.04%
B2
37.6%
30.0%
8,439
51.6%
76.3%
15.88%
Baa2
70.4%
58.6%
673
77.5%
76.3%
2.68%
Ba1
41.8%
30.1%
8,140
61.2%
86.3%
11.98%
A2
71.2%
64.7%
449
82.9%
86.3%
2.11%
Baa2
27.4%
21.5%
2,944
41.3%
17.6%
15.43%
A3
61.4%
51.2%
1,113
35.6%
39.0%
9.05%
Baa1
56.9%
52.4%
35
79.5%
81.7%
0.13%
Ba3
12.8%
6.7%
10
90.1%
46.3%
0.28%
Ba1
25.3%
18.6%
US ABS ex HEL
37
85.5%
78.8%
0.20%
Ba1
7.1%
4.7%
US CMBS
35
91.2%
88.5%
0.02%
B2
13.4%
5.8%
690
71.6%
89.4%
3.79%
Baa2
19.1%
16.2%
77
91.0%
89.4%
0.33%
Ba1
6.5%
4.7%
586
70.8%
93.2%
2.17%
A3
10.9%
8.9%
33
88.5%
93.2%
0.09%
Baa3
3.6%
1.7%
350
25.8%
52.2%
5.91%
A3
20.2%
15.7%
85
46.0%
66.4%
1.13%
Baa2
13.1%
8.8%
3
80.8%
83.5%
0.02%
Ba3
5.1%
2.0%
3
63.6%
56.8%
0.62%
Baa2
6.8%
3.2%
Other SF (includes SIVs)
Historical Averages Since 1993
US HEL (includes subprime)
excl '05-'07 vintages
US RMBS (includes Alt-A)
excl '05-'07 vintages
Global COOs
excl SF COOs
Int'l SF ex COO
a Other SF
Other SF (includes SIVs)
~~~~~~~~~:~:::~~~~~~~~~·.·December 2009
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Introduction
In a Special Comment published in April 2003, Moody's developed a set of metrics to measure the
performance of corporate ratings with respect to the dual objectives of rating accuracy and rating stability.2
Moody's corporate rating performance report is now updated on a quarterly basis.3 Moody's first introduced
and examined its structured finance rating performance metrics in a September 2004 Special Comment
"Default & Loss Rates of Structured Finance Securities: 1993-2003," and published these performance metrics
in a stand-alone document for the first time in September 2005. The structured finance rating performance
report is now updated on a semi-annual basis.4
For both the corporate and structured finance rating performance reports, the basic unit of observation is a
monthly cohort of ratings, i.e. all outstanding ratings at the beginning of a month are recorded and their
performance tracked over different time horizons. However, in computing rating performance metrics for
structured finance, Moody's incorporates both the default and loss severity experience of all structured finance
tranches because Moody's structured finance ratings rank order expected loss rates. This is in contrast to the
performance metrics published in the corporate rating performance report, which make no reference to loss
severity.
The most important measure of rating accuracy is the accuracy ratio, which for structured finance measures
5
the relationship between tranche ratings and their realized loss rates. This metric measures the quality of
Moody's ratings as indicators of relative expected credit loss risk.
As discussed in the April 2003 Special Comment, although relative rating accuracy is our primary objective,
Moody's recognizes that many investors are also concerned with the cardinal accuracy of the rating system.
In particular, they expect that investment-grade credits should rarely suffer credit losses and impaired
securities should normally carry low ratings well in advance of impairment. For this purpose, we regularly
track investment-grade loss rates and the average rating of securities during the 36 months prior to
impairment. Both of these measures should be low if the rating system is accurate in a cardinal sense.
We employ two measures of rating stability (or rating volatility) - the rating action rate (the frequency of rating
changes) and the large rating action rate (the frequency of rating changes of three notches or more) over a
twelve-month period. To examine how watchlist actions are used, we also report the percentages of
downgrades and upgrades preceded by watchlist (review) actions in the same direction.
See ".M.~.§§.l,l.rLlJ.g.1b~..E'.§.rfQ!.!ll.§D.9.§..9.LQ.9.r.l2.Q!.§J:.§.J29.09.B§.tlO9.?..." Moody's Special Comment, April 2003.
For the latest performance report, see "Ib.§.f.'.§rf9.r.m§.of.~.. QLIY.I.Qg.gy~§.. gQ!'p'.9.r.?t~.. Q~.RLB.§~.og.?:.. §~.I2.t.~.m.R§L;?QQ.~..Q\-!§.d:.§rJyJ'!p..C;!§J:.§," Moody's
Special Comment, October 2009.
4 Note that the criteria used to create the data set for this report have changed from prior performance studies. The most notable changes
are that pari passu tranches are no longer collapsed and wrapped tranches are included. For a more detailed description of the data sample,
~Iease see the Appendix.
The required adjustments to convert the standard default-based AR measure to a loss-based measure are discussed in :R.§f.?\-!JJ:..~..kg.§§
Rates of Structured Finance Securities: 1993-2003," Moody's Special Comment, September 2004. The concept of the accuracy ratio is also
described in the glossary at the end of this report.
2
3
~~~~~~~~~:~::~~~~~~~~~·.·.December 2009
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Accuracy Ratios
Figure 3 depicts the one- and five-year accuracy ratios over time with the following notable observations:
::
For global structured finance, the one-year accuracy ratio declined to 57.4% from 59.7% six months ago.
Excluding US HEL and US RMBS securities issued between 2005 and 2007, SF COOs, and the Other SF
sector raised the one-year accuracy ratio by 5.4 percentage points to 62.8%.
:.:.
One-year accuracy ratios declined on a year-over-year basis for all sectors except global COOs and Other
SF. US RMBS saw the steepest percent decline from its level 12 months ago, while the US CMBS
accuracy ratio changed the least over the same time period.
::
US ABS, excluding HEL, and US CMBS both maintained one-year accuracy ratios over 85%. If
transactions that closed between 2005 and 2007 were excluded from US RMBS, its one-year accuracy
ratio for the cohort ending June 2009 was 82.9%.
':':
The five-year accuracy ratio continued to fall for global structured finance, US HEL, US RMBS, and global
COOs. Further declines are expected for this statistic because it demonstrates performance on a lagged
basis and the effects of the recent growth in material impairments have not yet been fully incorporated.
Figure 3: One-Year (yel/ow line) and Five-Year (blue line) Accuracy Ratios
Global Structured Finance
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
~~;Q.;:::-:.::.-:~~/.~<_::~:.:. ~ ~.~.:~:.-+------....:.·""'······ ..·.:::::3.·::;.:.:::~·+----------~~-~~
\
+--------------+---------------+---------------+---------------+---------------+-~~~~~~~~~~~~~
,v;
\('
:);
:);
c::.:
(:
..:c·
:; ~
:; ~
:.:.
:.:.
I::
.::
C·)
r..___
.~~,
Global Structured Finance ex SF COOs, other SF,
100%
;(!5-'OZllintage US He. & RMBS
~~~ p,C~~:;S:\/:';::R.:::. • ::.:::iL:::{:.·:~···-··70% +-------------~60% +-------------~'
50% + - - - - - - - - - - - - - - - 40% + - - - - - - - - - - - - - - - 30% + - - - - - - - - - - - - - - - 20% + - - - - - - - - - - - - - - - 10% + - - - - - - - - - - - - - - - 0% +-~~~~~~~~~~~~~
.)
:()
.... -.
."
~S
,"»
-:»
",
",
(.~.
~
I::
':"
._)
c
,"0
.. :
Cohort Starting Date
~
"
0
~
0
.::
."
:,.
~
Cohort Starting Date
USABSex He.
USCMBS
100% ,------;---~~--:_---__;_;_;____cc__90%
~.='::::'"'.>.'-......-...----"':~
80%
:~:'--'
70% r-----i""':--'"'--'- - - - - ' - - - - - - - 60% 1--. :
50% :.' ::
f--J
....................: . }:.::.:.
f--}-.::f··} '.' .
------
~~~ +-·: _:";~.: : ~i_ _ _ _ _ _ _ _ _ _ __
:--
,"»
~
~-.-.•
'C.
'.'
.::
~
:,.
~
Cohort Starting Date
~,-.
I::
':"
.,
c
~
20%
10%
0%
+-------------+-------------+-~~~~~~~~~~~~~
.~~
(:
~,?
~ \;
:);
\
,,
..
-:'
;);
r..___
:);
\
I::
-
i~
-:'
(~ :
I::
..:c·
Cohort Starting Date
Note: At the beginning of each month, securities are grouped together by their alpha-numeric ratings to form rating cohorts.
The latest one- and five-year cohorts are formed on July 1, 2008 and July 1, 2004. Breaks in the accuracy ratio series occur
when the number of impairments for the cohort is less than or equal to one. Crosses in the one-year series and plus signs in
the five-year series indicate that the accuracy ratio was computed from only one impaired security.
~~~~~~~~~~~:::~~~~~~~~~·.·December 2009
::::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Figure 3 (continued): One-Year (yel/ow line) and Five-Year (blue line) Accuracy Ratios
USHB.
us HB. ex '05-'07 Vintages
1~~~ ~:::::: .. ,.::.:~} ........ :.::.:::. ::::;:;:2:,,/\ ":.~-
1~~~ ~:::::::::,:;.::>:'::.:;::,.,-:"'~"~'-~~'::+2:C~~""""-'_.. _
80% +-~~~~~~~~~~~--":c--::;-70% +-~~~~~~~~~~~~~~-\.~;60% +-~~~~~~~~~~~~~~~--:50% +-~~~~~~~~~~~~~~~~<.
40% +-~~~~~~~~~~~~~~~~-:.
30% +-~~~~~~~~~~~~~~~~20% +-~~~~~~~~~~~~~~~~10% +-~~~~~~~~~~~~~~~~0% +-~~~~~~~~~~~~~~~__~
80%
70%
60%
50%
40%
30%
20%
10%
0%
+-~~~~~~~~~~~~.~~~~
+-~~~~~~~~~~~~~~~~-+-~~~~~~~~~~~~~~~~-+-~~~~~~~~~~~~~~~~-+-~~~~~~~~~~~~~~~~--
+-~~~~~~~~~~~~~~~~-+-~~~~~~~~~~~~~~~~-+-~~~~~~~~~~~~~~~~-+-~
__
~~~
__
~~~
____
~~~~~~
:~
Cohort Starting Date
Cohort Starting Date
us RM BS ex '05-'07 Vintages
us RMBS
. ;:; .:;-.. ,.....·········'············;···:..=;·;,
cc·.,cc.c-;-=;--~----;~---c=~..
100% ~.:=
90% -----.:,,,.
z;-·~·-
\~
00%
70%
60%
50%
40%
30%
20%
10%
0%
+-~~~~~~~~~~~~~~~~:+-~~~~~~~~~~~~~~~-+-~~~~~~~~~~~~~~~+-~~~~~~~~~~~~~~~+-~~~~~~~~~~~~~~~+-~~~~~~~~~~~~~~~+-~~~~~~~~~~~~~~~+-~
______
~~~
__
~~~~
______
~
100%
90% ~:;(:;:::';.·:·.·......;;c;c;c;.;:;.,,:,.:::;:;cc,....,~.,:,.::.:" ...,; . ;/ .. :;~ . ;_
80%
70%
60%
50%
40%
30%
20%
10%
0%
+-~~~~~~~~~~~~~~~+-~~~~~~~~~~~~~~~+-~~~~~~~~~~~~~~~+-~~~~~~~~~~~~~~~+-~~~~~~~~~~~~~~~+-~~~~~~~~~~~~~~~-
+-~~~~~~~~~~~~~~~+-~~~~~~~~~~~~~~~+-~~
__~~__--~~~~~__--~~~
"-.
....
~
Cohort Starting Date
Cohort Starting Date
Global COOs
100%
90%
80%
Global COOs e x SF COOs
~~~~~~~~~~~~~_~~+-~~~~~~~~~~~~--:~~-
:.
."........--.;--;_. : -'~~
+---~.;:;:;.
70%·m~~
60%
50%
40%
·; ........ ,/.;···.:;-;c!:~~:c-··;··············'·... 7."'--'.;:.;-:'---~~-';--
~~;, :======~-;:l~-:~-:. .-. .-.:, -. .--.c:~ -;:7.-..-...,-....-./~./;·: .;:. ;~,
50%
+-~~~~~~~~~~~cc-~~-
40%
+-~~~~~~~~~~~~~~~~
+-~
______
~~~
__
~~~~
______
~
~~~~~~~~~~~~_~
600/0
+-~~~~~---~~~~-;,;-,~~~;.~-
~~~ :========-----7~;~~~~~~~~:~~~·;~c;-,-l
10%
0%
100%
30%
20%
10%
0%
........_..
+-~~~~~
7-~~~~~~~~~-
+-~~~~~~~~~~~~~~~+-~~
__~~__--~~~~~__--~~~
:~:
Cohort Starting Date
Cohort Starting Date
Int'I SF ex COO & other SF
.. .."..
;.::,:,:.:........ l:
other SF
+-~~~~~~~~~~~~~~~~--
100%
90%
80%
70%
60%
50%
40%
+-~~~~~~~~~~~~~~~~--
30%
+-~~~~~~~~~~~~~~~~--
20%
10%
0%
~~~~~,.o;o;="'.~ ~---c::;
f
'--7--;';
".
~~~~:_
·.c·.. ·.·
't::r
"';"ri~l--";:''-'- - - : c · ·
•
..:.,:
+-~~~~~~~~-;.;,;
cc-~~~~~~+-~~~~~~~~~~~~~~~~--
+-~~~~~~~~~~~~~~~~--
__
~~
__
~~~
____
~
__
----~~
~~~~~~~~~~~~~~~~-
..
+-~~~~~~~~~~~~~~~---c
+-~~~~~~~~~~~~~~~--c
+-~~~~~~~~~~~\~~~~~:+-~~~~~~~~~-
..-------::-..\.
l :."
.. . ' - - - -
~~::
+-~~~~~~~~~~~~~~~-
+-~~~~~~~~~~~~~~~+-~~~
__
~
__
~
____
~~~~~
".
.... -
~
..:c·
Cohort Starting Date
~~~~~~~~~:::::~~~~~~~fDecember 2009
co::;
c·
:.:.
:c·
+-~~
":~ ::.:-
+-~~~~~~~~~~~~~~~~
..) .
100%
",
90 IC
80%
70%
60%
50%
40%
30%
20%
10%
0%
;.;..;-
.~;.
..:c·
Cohort Starting Date
Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
__
~
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Investment-Grade Loss Rates
Figure 4 shows one-year and five-year investment-grade loss rates by sector. Note that:
::
The one-year investment-grade loss rate increased for global structured finance and all sub-sectors with the
exception of the Other SF category in comparison to the rate a year ago. However, it appears that the loss
rate peaked for the cohort ending March 2009 for global structured finance and has been steadily declining
since then. The one-year loss rate is reduced by 75% if US HEL and US RMBS from the 2005 to 2007
vintages, SF COOs, and Other SF are excluded from the calculation.
::
US ABS ex HEL, US CMBS, and International SF, excluding COOs and Other SF, experienced low oneyear investment-grade loss rates for the cohort ending June 2009 at 0.2%, 0.04%, and 0.1 %, respectively.
:.:.
Because the five-year investment-grade loss rate is a lagging indicator, it still did not show the type of
increase displayed by the one-year loss rate.
Figure 4: One-Year (yellow line) and Five-Year (blue line) Investment-Grade Loss Rates
Global Structured Finance
Global Structured Finance ex SF COOs, other SF,
3.0%
18%
16%
i I..I.if\obn'J.1'-L,,,LJ""-I"l':"ICL...I=IR.<>"-"Illl
RIIII1.c><:>.R!::_ _
~....i1ll
~nrl~...!J"n.".'i'_"...l.'ln L
7'-,,"r.ll.i·n
2.5% + - - - - - - - - - - - - - - - 0 .
14%
12%
10%
••
.,••
::::
8%
6%
4%
.........
::':'::
0.5%
~~~
:.~,
g~
~~
:i;
l~;
~~
,'.
:~-(
--:.
:D
l~;
l~;
0.0%
,):'
'.'
'..<.'
0.5%
..
0.8%
t:.
t:q(/
0.6%
:::
: :
;.:\
0.4%
0.2%
..••.;.....
.;/
~~
.,~
."
.::
~-------------
L:-·.----------'=·:L·r: ='= = ='='i.~.- 1J:~=J!·-t-L.-."._-._;_._-.-,.-,-..
:}\~
,.-c-.-;:;
.: ~.(
0.0%
~
1.0%
~~
;.
1.0%
:".-:
,-,;
USCMBS
:-:~
1.5%
~~
:';..1
Cohort Starting Date
./...;>.,\:
;;.
,
=-;'
c.
1.2%
}
~~)
.::
3.0%
2.0%
:'1
. ;:
........................ ,.~
~
HB.
US ABS ex
.....
t::·::··:.... r··-r:-,:::·1·~~·1·
Cohort Starting Date
2.5%
.. =-.....,,_ _--'.:...:...:.l_____:_
;.~
••:.•.. •.=.•".;•..•.!/.:".':.y.;
..-'
r,~
.;.-,
1.5% + - - - - - - - - - - - - - - : 1.0% +------~/:7":=:...... .
:'
.:
2%
0%
2.0% + - - - - - - - - - - - - - - . ; : .
~.
(.~.
,-"-,
.;/
'.'
(-~
.,~
."
.::
I::
'e.
':"
Cohort Starting Date
I;)
b
.,~
...'l:
c
h
0.0%
~:" .
~~
'e.
."
.::
....
~
-.
~~
:))
-:»
c
...'l:
2-
c
...'l:
.;.-.
2-
Cohort Starting Date
Note: At the beginning of each month, all securities carrying an investment-grade rating are grouped together to form a
rating cohort. The latest one- and five-year cohorts are formed on July 1, 2008 and July 1, 2004.
~~~~~~~~~~;~:~~~~~~~~~·.·.December 2009
::::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Figure 4 (continued): One-Year (yellow line) and Five-Year (blue line) Investment-Grade Loss Rates
30%
us HB. ex '05-'07 Vintages
5%,------------------------------
25%
4%+------------------------------
USHB.
20%
>
15%
10%
}
5%
I·":':"
0%
,-,-.
....
~
.;.-:
,y~
-.
;~
~.~
....
~
"
-.
')
;.:;.
l~;
l~;
'.
;~
:~:
:~:
L
'.'
v,
~~~
t~:
'~F
::'~
.S;
.S;
:S'
.;.-:
,y~
Cohort starting Date
'.';
~
~
cC·)
T~
"
"j
Cohort starting Date
us RMBS ex '05-'07 Vintages
2.5% ,-----------------------------
USRMBS
20%
18%
16%
14%
12%
10%
8%
6%
4%
2%
0%
,\
2.0% +----------------------------,',
"
1.5%
+---------------------------~,-
'.
,.'
1.0% +----------------------------'••"
-------0;.,:..'
0.5% +--------------------.----,-......
.,'
,-,-..
i-\~
.;.-:
.;.-:
,y~
,y~
-.
c....-.
~
;~
;~
;~
~.~
....
~
~
2~~
~~
,'.
:~-(
"
0.0% -,·.··c···'(··.····,··,·,.,··,····.·,·,·,·,,···,····.···.,....,....".,',.,.•...,..,.,.,.",.,{::,...,.... ,.
,'.
'"
"'-.:
:~-(
Cohort starting Date
"',.,
~~
"
-~~
Cohort starting Date
Global COOs
Global COOs ex SF COOs
10%,-----------------------------
25%
9%+-----------------------------
~~~=============================.:.
20%
15%
6%+----------------------------:-
10%
~~ +-----t-d'~:'·~\i'~·\·.\,:. -----------+,:
5%
2%/'
0%
1%
0%
'"
"'-.:
~~
"'-.:
:....
r;;.
r~>
\",'~
r~>
-~~
-~~
"
c.:.,
-~~
-~~
-~:)
,", ....
~::\
"
~n
;))
-:.:
-~:)
.~j
")
,',',
1"-.
C~
:S'
Cohort starting Date
Cohort starting Date
Int'I SFexCoo & other SF
0.20% ,----------------------------
6%,------------------------------
:::
:~ ~ ~ ~ ~ ~ ~-
4%T--------------------.(~~
:J
'
':':'~"
,.L
::,.:L--:J,
.
:i.,!;
.. .....
.............
-.,
.....
;',j
~~
::':-j
other SF
5%+----------------------------,-
0.00% .. ,.. ,..,.. ,., .. ,.. ,., .. ,., ..,.. ,.... , ..,
"
.':)
;~;
~~
Cohort Starting Date
.~.~.~.~.:;:::.~.~.~.f·December 2009
",)
,'.
--------c:.'--.,
. 'd\-:----;.:•.-'
·~:i:::,-.----.
.......,. .::..,:-l-----:'-----r"-.T.-.;.-.r-'-.-.-~-.-r., .. 1 ., .. ~~~.... ::
., ..~
3%+---------------------------.:--
,:' .'
2%+-----------------------------1%
::....
O%+«~~~~~~~~~~f~~(J
..
::~
::,
(:
~,?
Cohort starting Date
::::, Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
\('
j~,
::1
(:
\~)
.. )
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Average Rating Before Impairment
Figure 5 presents the 36-month-average rating before impairment over time as well as the number of newly
impaired securities used to calculate the average ratings. The following observations are noteworthy:
::
The 36-month average rating before impairment for global structured finance was Baa1 for the most recent
cohort, one notch above the average of Baa2 six months and twelve months prior. Excluding SF COOs,
the Other SF category, and 2005 to 2007 vintage US HEL and US RMBS, the average rating before
impairment dropped one notch to Baa2.
::
US RMBS and global COOs exhibited the highest average rating prior to impairment of A2 and A3,
respectively. For the latest cohort, the 36-month average rating for US HEL was also high at Baa2.
::
All other sectors displayed average ratings before impairment that were below investment-grade with US
CMBS experiencing the lowest average rating of B2.
Figure 5: 36-Month-Average Ratings before Impairment (yel/ow line) and Number of
Newly Impaired Securities (blue line)
Global Structured Finance
Global Structured Finance ex SF COOs, other SF,
A a 3 , - - - - - - - - - - - - - - - - , 25000
A3 ~-.----------~:!~ 20000
Baa3
15000
Baa3
-~---~.----c.----c.J~
10000
Ba3
t-------------:;-:;
,::
--t 5000
(;aa3 -, .. ~.r.'.' ..,..,.. r.I.' ..,..,.. ~.,., ..,..i·-·-~-··~·,-''-'',;.!.;,'--r'--ro,o·,·_·.,L 0
'-'-.,
(~
.
;~:
',.'
~:.
;(,
,:'1
~~;
(;,
,~:
;~:
(:
C·)
~:.
;(,
3000
A3+-~-------------~~2400
~.:-:.~.--~---~.::..~.. --~!
Ba3 r------"-:<::... ,"/:"-"....-.
B3
Aa3
1200
B3+-_ _ _ _ _ _ _ _ _ _
Caa3
'-'
C·)
,~:
c
>::
':,
Cohort Starting Date
7~ t---~;-------c'Trl\
i ::
B3t-------.,.-.):---;······:~·--vJ
,
"
....:
~
c:
,~
Cohort Starting Date
c::. ~
.-
.::
':,
Aa3,-----------------,500
A3
~~
-,
t
Baa3
t
t-------------~:lt... :
'"
400
300
~:: _ _ _ _ _ _::::+ 200
B3 t-------c---:""'-:~ ~
100
Ba3 +-_ _ _ _ _ _
Caa3 .........( ........,.. , .........., ... , ... , ... , ... , ... 0
:.
\!",
~:)
..
.,
...... ,
,..~>
,..~>
."
c(
,<)
u;
USCMBS
USABSex HB.
,<)
>::
Cohort Starting Date
Aa3,---------------,200
t
r..___
G.
,:J
(:
~~-:~j 600
(;aa3
. .-.J
·,·~·'··,··I·,··,··r·'··~··r·r"··-·······
,j)
")::
:))
(":
,
.~,
..
..:c·
...................... "-'.-.'~
r:
,~
;
0
:;)
':-;
i::
,
..C'J
Cohort Starting Date
Note: At the beginning of each month, all securities that become impaired within the next 12-month period are grouped to
form a cohort. The latest 12-month cohort is formed on July 1, 2008. Breaks in the average rating before impairment series
occur when the number of impairments for the cohort is less than or equal to one. Crosses indicate that the average rating
was computed from only one impaired security.
~~~~~~~~~:i:;:~~~~~~~~~·.·December 2009
::::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Figure 5 (continued): 36-Month-Average Ratings before Impairment (yel/ow line) and
Number of Newly Impaired Securities (blue line)
us HB. ex '05-'07 Vintages
USHB.
Aa3
10000
A3
8000
Baa3
6000
Ba3
B3
2000
B3
"-
!s'
cC·)
._)
~~)
;)~
~,
"-
:s'
::'~
.)
C·)
!s'
.. ,/
l
::/
..
~
....-",
0
'C.
Cohort Starting Date
"I::.
'-.... -
I:'
---.::
."
:,.
450
300
150
_,ff
--,
~
::'~
._)
:,:}
'-:.'
~;~
,~,
C
"-
.S;
'\:
-,//:
Caa3
0
,-
600
{
Baa3
4000
:'1
!,
A3
Ba3
Caa3
750
Aa3
o
,-,-.
~
Cohort Starting Date
us RM BS ex '05-'07 Vintages
us RMBS
Aa3-,---------------------,1oooo
A3+-------------~~
Aa3-,------------------,500
A3+--------------~~400
Baa3 +-_::"------c-----c:--:::c-:.c-----.~:~H 6000
Baa3 +-----:,::f-------::---,'.c--c--------:=t 300
Ba3 +---:.----c'"-----':~:~---".,'.-'-----~::'--------'-'c .. :-~:_t 4000
B3 t-------------:~ 2000
Caa3
"I· ., •• ,•• ~ ",., ••, ••
l!";
:'1
cC·)
e.;;
~
o
r ","\ ••,•• ,•• ~ ",.,., ••,•• ,•• r .,"\ ••,•• ,•• ,"\ ••,•• ,......
,-
~~)
;)~
Caa3
-C 'T' " " " " " " ' " ,-,"',-" ","
.)
G~
cC·)
~
,~
;
"
:):
i::
C'J
+-----~----= c-¢ : :
B:: +-__________....-:.::). ~t
.:
(;aa3 , .. ~., ...... l . ' .... I.l .... r.I.' .• ~.~,'--'••·----·-·-·c.'····-","'"0=--'=--=--,=,,-<_~
__ ~
___
,---i:- - - - - ' -
..
,-,j
;))
,:.:
;-
1400
700
0
A3
Caa3 +""
.~j
C'J
C'J
--j
--j
:':'
PP'f",', 'P'f' ','"
cc~
:c'
Cohort Starting Date
Inti SF ex COO and other SF
A3+---------------~~------------+40
~~,n=:j::
~~
d
:~;.:
--,
:~~:
,:.:
'~j
Cohort Starting Date
~frjrr~f--'December 2009
:::::::::::::::::::::::: ..
--
:S'
1000
,~
;
""r"'-"Li'---,-~___,_;_"",,'-"f"",'-','-'-,,'-f_'r__r+ 0
e.;;
c·
c~
,~
;
(:
(:
C,)
.. )
C,)
.. )
Cohort Starting Date
Aa3,-----------------------------,1 50
,~:
o
~
B3+-----------~--~_~250
.~j
~~
::~
,-"'-,-,-"',/~
Ba3 +---------: ,-:-" ""--"._-----'-----':+ 500
'.'.'
--
I::
(":
!~
Baa3 +-----~-------.:~~t 750
,:.:
~~;
:0
t-----c:::::~---------k
,:J
(:
--
c:
Global COOs ex SF COOs
Aa3 - , - - - - - - - - - - - - - - - - - , 1250
,:J
(:
I
-,--
Cohort Starting Date
Global COOs
Aa3 - , - - - - - - - - - - - - - - - - - - , r 3500
;))
\
::'
Cohort Starting Date
Ba::
,-" ,-,-"'" , - , - " " " ' " ,-,-,-,-,-,--'.'-"',
;J:
:;)
-,
":';..1
,',;
B3+--------------:-+100
I
--
~~
other SF
Aa3,-----------------------------, 50
:}".:.
A3+---------------------------~·c-f
1':,
40
Baa3+-------------------------~~~+
30
i'l,
'\
B3+-_________________________I
10
Ba3+-________________________~~~:~ 20
~__+,
i
:0
...;;,
c
,0;;'
._)
~-..-.
~~
\;.'
:;:;
,-,
(r·
::.~:
c
~~
:i;
,0;;'
._)
Cohort Starting Date
::::: Special Comment:::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
,-,
c
,0;;'
._)
0
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Rating Action Rates and Large Rating Action Rates
Figure 6 reports 12-month rating action rates and large rating action rates. Figures 7 and 8 further
disaggregate rating actions into downgrades and upgrades, and Figure 9 demonstrates how frequently
downgrades and upgrades have been preceded by watch list actions in the same direction. 6 Key observations
include:
:':'
The one-year rating action rate rose to 58.3%, a 61 % increase from the rate of 36.2% six months prior and
2.5 times larger than the rate of 23.6% a year ago. The large rating action rate was 49.9% indicating that
most of the rating actions involved movements of three notches or more. Excluding SF COOs, the Other
SF category, and US HEL and RMBS transactions that closed between 2005 and 2007, the overall and
large rating actions rates declined to 33.9% and 26.8%, respectively.
:':'
As has been the case since early 2007, downgrades dominated rating changes for the latest cohort as the
12-month global structured finance downgrade rate rose to 57.5% while the upgrade rate stood at 0.8%.
Downgrade rates increased compared to their levels six months ago for all sectors and US ABS ex HEL
was the only sector that experienced a significant increase in the frequency of upgrades.
':':
The proportion of downgrades that were placed on review prior to the rating action increased slightly to
45.8% for the cohort ending June 2009 versus 44.3% for the cohort ending December 2008. The increase
was caused by a rise in the frequency of reviewed downgrades among US HEL and US CMBS as most
other sectors experienced declines in this rate.
:':'
The percentage of upgrades that were placed on review prior to the rating change fell to 4.1 % for the July
2008 cohort from 8.0% for the January 2008 cohort. Declines were also seen across all sectors except
US ABS ex HEL and US CMBS.
Moody's also reviewed its uses of watchlist actions on structured finance securities in a Special Comment, "Structured Finance Watchlist
Resolutions: 1992-2003," June 2004.
6
~~~~~~~:::~~~:::~~~~~~~·.·.December 2009
:':': Special Comment ':':' Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Figure 6: 12-month Rating Action Rates (yel/ow line) and 12-month Large (three
notches or more) Rating Action Rates (blue line)
Global Structured Finance
Global Structured Finance ex SF COOs, other SF,
20%
70%
,-~~~~~~~~DC~~~~--~c
60%
50%
r
40%
:i
30%
10% +-----------------------------:~~
:;t
20%
}
10%
0%
15% +-----------------------------~~
.:.,.,:: ........
i··,·.;::-:-·:······::·······. .:.,.,.
,,",
'-~:
..
...-.-,
';}
'-~:
...~>
~;j
.-..~
c:
,~
..
.-..~
ji
..
~~
-,
c:
.0
,~
,~:
;:.\
Cohort Starting Date
;:.\
~~
5:
1
.' .
~~
,~:
;:)
,:.:
.~j
USABSex HB.
USCMBS
30% ,--------------------------------
40% ,-------------------------------
25%
+-------------------------------.:;:-
20%
+-----------------------------~p:
15% +--------------------;::~L/·
.:'
1:: +--------------------"l0W
0% ~~::;~;-:;._~:{..,,:::!'r-·i;·::~;:·-T·-·-·"-:·:·:"r"::·~:~::.,.;:~:~.>~~~~~_T_c~
;.:~
-.
... -.
.',
~::~
36% +-----------------------------~
32% +------------------------------:
'",:)
.......
c
,0;;'
.. )
.
....
~~
~~
:i;
c
c
.. )
.. )
,0;;'
,0;;'
~~
\;.'
:i;
:;;:;
:0
;))
'",:)
'",:)
c
,0;;'
.. )
Cohort Starting Date
:i;
c
,0;;'
.. )
~::~
,"
~~
c.6:,:d-
:i;
,'.
~~
:~-(
I;)
.,
~
",.'l
Cohort Starting Date
US HB. ex '05-'07 Vintages
USHB.
75% ,--------------------------------
50%
60% +-----------------------------~.
40%
::: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~J~f--·~
30%
20%
10%
~
0% .• ;........ &;••• :.;:·"'·:.·•.,•. ?::::·:·:i:·::··•••• •••....•..••.••.•••.•.•.••.:-:...:.:.:....... :.)
~"
(~.
.S;
.S;
-,
~)
~.
,...)
;.:1
C
t
t
_~
Cohort Starting Date
Special Comment
6d
!j~~~ ,j~Tr~;~;;.;:.
,4~~. . ....:.: . . . . . . . . .
;-")
co:::
.0
Cohort Starting Date
:<.
~~~~~~~:::~!:i:~~~~~f·December 2009
..
~;:
c(
co:::
0%
~;~
",
~
...-.-,
'C.
c.'
.::
."
~-".'
'C.
:(
~
'.'
.::
."
Cohort Starting Date
Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
c
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Figure 6 (continued): 12-month Rating Action Rates (yel/ow line) and 12-month Large
(three notches or more) Rating Action Rates (blue line)
us RM BS ex '05-'07 Vintages
us RMBS
80%,------------------------------
30%
+------------------------------
25%
70%
~~~+-----------------------------(
20%
+-----------------------------~
15%
+----------------------------~j-.
10% +---------------------------~j:~{-
10%
40%
30%
+-__________________________~:·ri
20%
00/0 L:.•::.:;.-,.,.,-_,••_ l········r:;:·/-::::::;,·;:·<;········'·:·,···
~~~
~~~
~~
~~
r- ••.•.•.•.•:.-•....••:.:.;.::.• "',. .•~-
0%
:D
:i;
:i;
:i;
:i;
5%
,
l!";
:'1
C
~~;
C,)
e.;;
~
Cohort Starting Date
:'1
C
~
C,)
,
~~
:",-:
;;:;
~
Cohort Starting Date
Global COOs
Global COOs e x SF COOs
70% ,------------------------------
60% -,------------------------------
60%
+_--------------------------~'
:~~ :============================--J't-['
:::
30% :============================---ii'
+_----------------------------~
30%
20%
10%
00/0
+-______________
~.2, ..---------~~
-----c,:~~----.. 7C.:~:.:~:~:::,-····\ .... : :.: :.: .: . :.:.::.jL
..,..
.,.·.~.'
·r ••,•• r.·.,.:·~:.-;-····-~:'-···¥'·-·-·~-··.•';-'
.•••~.~~~~~~~~
.... -.
c"
~:)
,..~>
\!",
~
t
c:
.':)
,~
:~
t
"',:
c(
-~~
20%
10%
r--:'~,-----.. / . -.--.' .~:f
....~ . . . . . . . . :.:~;;-.l
..
..:
'.:\.:.
. . .:. . . . :. :::.::.::: ............
0% .,.. ,.:::: ... ,.:•.....,...;... ,:..';. .;'T'~~~~-,;,,;c.;--c~-T-r~
~-:\
(;
)::
.':)
Cohort Starting Date
I~'
:.:.
',',
")::
..:c·
:~
r:
(":
,
.~,
..
,~
;
:;)
':-;
i::
..,C·)
Cohort Starting Date
Int'I SF ex COO & other SF
other SF
20% ,------------------------------
30%
16%
.' : ; - - - - - - - - - - - - - - - - - - - - - - - - - - - -
25%
: 1i
20%
12%
+
.....:.r-
..;.....
.:-:--------------------------~
15%
8%
;/
4%
0%
.
' . ' .'.
}:":";
'.
...~cc·---~'CT~
....
~·.L.~.,. :L",.L::·:;·;;·:::::t. ;. ::~.:~~:;;)-·-·-·····/,:/· . .{. . .-~: ·u.:._••_.__.~._·
\"'.',
C,-~~
.... -.
c"
~:)
,..~>
.':)
\;,
~
c:
:~
t
,~
t
'-.:
Cohort Starting Date
c(
-~~
~-:\
.':)
10%
5%
0%
(;
I::
',',
..:c·
!~
r:
(":
,
.~,
..
,~
;
:;)
':-:.
)"-0.
r:
i::
,
C·)
..
,~
;
Cohort Starting Date
Note: Rating actions include upgrades and downgrades, which are measured on the alpha-numeric (or modified) rating
scale. Rating cohorts are formed each month covering a 12-month period. The latest 12-month cohort is formed on July 1,
2008.
~~~~~~~:::~~::~::~~~~~~·.·December 2009
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Figure 7: 12-month Downgrade Rates (yel/ow line) and 12-month Large (three
notches or more) Downgrade Rates (blue line)
Global Structured Finance
70% ,-------------------------------
Global Structured Finance ex SF COOs, other SF,
40%
amI '05_'07 \/intanp US I-II'L & RM BS
+-------------------------------
35%
:~~ :=============================~;r
30%
60%
30%
~1~ :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~-1·
+-----------------------------~~
20%
./
1~
)
00/0 ,.·~.·~.·.r··4_·······~·91 .... I ••········;;·:·;;··;-:-:;:;:·;·i····r.~
\"'.',
~:)
,..~>
~~
,<)
C,-,
~-.
."
t
...... ,
c-.
-,
-,
...
....:
,<)
~
~
....:
+------------------------------+------------------------------
5%
00/0
.. ~....
!;-;•••:•••.•:.:.:.;.;'.'••;.....
'-'-.,
(~
.
;~:
15:
,,-;
.
;~:
;:)
~:.
;(,
c;
,:.:
,:J
(:
.~j
C,)
Cohort Starting Date
Cohort Starting Date
USABSex HB.
30% ,--------------------------------
USCMBS
40% ,-------------------------------
25%
35%
+-------------------------------i.
+------------------------------l.l
15%
+---------------------~------,~
.::::. ~./..
l
+ -., :.-:.~ - ., _- ; ,._-: ~-.; -;: - :-. - -:.;;;-r,J
---_y
__-
....'.,
:~-(
'-;:.
:i;
~~;
---.::
,y~
---.::
,y~
~~~
~~ 7···'tt::. t.;:'*"'~' ....." .: ~": ~: .":.~~'~. .:-.:---:-::.:-:-:-.
,-:~-(
-------------------------f:
+-----.
1
•.•: .
.•.•
--._:.;-.:
•••
'0
+-------------------------------
30% +-----------------------------~:
20%
'::
:-.::;-:,l-
."
f
:;;:;
\!",
c-'
,~~
Cohort Starting Date
70%
+-----------------------------~.
+-------------------------------"
..
'
;~~ :+-==_-==_-==_-==_-==_-==_-==_-==_-==_-==_-==_-==_-==_-==_-=: , . f~=·)t
10%
+-------------------------------c.,;~:-
+-__________________________~~r~
0% --,
c···l·:···~r···~~··;""r··i··t. ~··'·9--~~T~
2 ~ ~
r
~
~
I::
I::
I::
.::
.::
." ." ':' ':' ':'
..
Cohort Starting Date
."
f
...... ,
c- •
"',:
,':)
C"'
-,
~
us HB. ex '05-'07 Vintages
,-----------------------------~
60%
20%
~"
Cohort Starting Date
USHB.
80%
~~~~~~~:::~~:::::~~~~~~·.·December 2009
(~
~~
.,>{
• ........................:•••;;:.:';.•••.•••••••..••••
60% ,------------------------------50%
+-------------------------------
40%
+-------------------------------
30%
+-----------------------------~
20%
+-----------------------------------:/',::1
10%
+---------------------------~~-.....................................;..::/
;;
.';
;:,~
:,";"
j~:
~~
j~:
,.,-.
:)
:..
.:-:
:~
;:;j
Cohort Starting Date
::::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Figure 7 (continued): 12-month Downgrade Rates (yel/ow line) and 12-month Large
(three notches or more) Downgrade Rates (blue line)
us RM BS ex '05-'07 Vintages
us RMBS
80%
~---------------
70%
+----------------
:
I
)
:
30%
25%
20%
15%
10%
20% +----------------~-
1~
J
0% ........................ ····c··,··r-¥-·c··,··,··,··,··,··,··,··,··,··,···M··'··,·., ..~
...
"'.
'-~:
'-~:
".-.-~
".-.-~
:}
5%
,.::'::.:::.
...
0%
,.
'< ..
';}
(-~~
(-~~
G~
-~~
-~~
-~~
,~
::.
;
Cohort Starting Date
Global COOs
60%
+-______________::c
-:~-:-
+-----:-.-.:_
... __. -...-}-:::::-::.-:...
r"-' r'
j"
r "1-" '1"
...
:
::'
C·)
..
"
i::
,
..
..-:' -
.:'1
..
..
i::
:; ~
:; ~
"
,
1'
L
-·
..-.. :-.::.:-:::::-::.::.-l-i
::.
20%
:; ~
:.:.
!:3
(":
,
.~,
..
+--------:::::::...
::J
+----~ _______::./-----'i~i
.,.---,---,-----,.;.:...........-.-.-......
0% ,·-·"0"--.----,
:))
".:)
c
:i;
,-;;,
\>.
"'r o -::::::':':':'::-:)'
,"
~~
:i;
~::~
. .:.1
~~
:~-(
")
Cohort Starting Date
Int'I SF ex COO & other SF
other SF
14%
~---------------
12%
+----------------
10%
))c-:~- - - - - - - - - - - - - .
8%
30% , - - - - - - - - - - - - - - - -
25%
20%
15%
10%
+--------------~:./
+----------------
+-______________;'\~C
f-C
::
5%
'', - ... ......,:.:...;....•..,-;--:~,~-~~-:::-,---{
•.,
O'IC,
~ :.,:
:
"
i::
::'
I;':'
:: +-----------------;f:
Cohort Starting Date
...
:.:.
..
60% , - - - - - - - - - - - - - - - -
10%
.
>.;~'.~-.- -:-.-:::.:.~
C·)
..
:; ~
~
Global COOs e x SF COOs
~---------------
00/0 ,"
~\
Cohort Starting Date
70%
~[~
.J
,:):
,
C·)
..
.:'1
..
..
~~~
i::
:; ~
:; ~
:; ~
"
,
C·)
..
Cohort Starting Date
'~
:))
".:)
::'
:i;
c
,-;;,
,"
~~
:i;
~::~
. .:.1
~~
:~-(
")
Cohort Starting Date
Note: Downgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month
covering a 12-month period. The latest 12-month cohort is formed on July 1, 2008.
~~~~~~~:::~~~:~::~~~~~~·.·December 2009
::::: Special Comment ::::: Moody's Credit Policy -- The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Figure 8: 12-month Upgrade Rates (yel/ow line) and 12-month Large (three notches or
more) Upgrade Rates (blue line)
Global Structured Finance
Global Structured Finance ex SF COOs, other SF,
5%,-----------------------------\: ':--------------~: ...,:c-",----......: ..
4%
"--
~,
...... ,
~,;
...... ,
...... ,
'-,
~5
'-.:
'-.:
'-.:
'-.:
,<)
,<)
")
~-:\
,<)
,<)
4%
+----
3%
::-
::~~
,,-;
I..!-,;
,,",
c,'
c,'
c,'
Cohort Starting Date
c.'
c,'
c;
,,-;
c;
I..!-,;
c;
1-",
c;
Cohort Starting Date
USABSex HB.
USCMBS
5% ,-------------------------------
18% ,------------------------------
4% +------------------------------3%
+-;~,,:-----------------------:c:---.
2%+-:--;--------0:<---:~----~~.--~~~
16% +-----------------------~.:":---14% +---------------------------:,-:':;-,-;-',,12% +-----------,-----------------:','-,"-:---.,:-.
10% +---_:~~_c_------------::----
!~~~~:
";:.
~~~
~~~
~~;
~~
~~
Cohort Starting Date
-.,
--
--:.
~-~~
:~-(
~~;
~~;
.::
.....
,y~
,--
:~-(
Cohort Starting Date
US HB. ex '05-'07 Vintages
USHB.
5% ,-------------------------------
5%
4% +-------------------------------
4%
3% +-------------------------------
3%
2%+-------------------------------
2%
1%
::'~
::'~
0%
:'1
."
:'1
OJ
.)
C
C
C
C
C,)
")
C,)
")
C,)
")
C,)
")
Cohort Starting Date
~~~~~~~:::~~::~:;~~~~~~-_-December 2009
~;~
--,
~
~::-~
--,
~
~»
:))
--,
,'C)
c
,'C)
-:»
c
--,
.;.-.
2-
2-
Cohort Starting Date
::::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
2-
I::
':'
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Figure 8 (continued): 12-month Upgrade Rates (yel/ow line) and 12-month Large
(three notches or more) Upgrade Rates (blue line)
USRMBS
us RMBS ex '05-'07 Vintages
7%,------------------------------
7%,------------------------------
6%+------------------------------
6%+------------------------------
5%
>:':.
.'--------:,------+-------c------.
5% +-------:.~.-----
.--------- - - - -
;~~~~ l~~~~
4%
,,",
c
,
I
c.'
c,
;:)
I
,:.:
d
+-----~• L______--o.
?£?
,:.:
-.,
.~j
.~j
-'j
-'j
-.,
.....
;.'
,:.:
.~j
-'j
.....
:~-(
:;;:;
:~-(
Cohort Starting Date
if
...
.::
.::
,y~
. .:.1
:~-(
,y~
Cohort Starting Date
Global COOs
Global COOs ex SF COOs
14%,------------------------------
14% ,------------------------------
12%
12%
/
---------------------------
~: :
10%
8%
6%
;'
--------------------------
8%
6%
)
4%
..
. - .- - - - - - - - - - - - - - - - - - -
2%
.-:
.
. ...,.....•.
0% .. ~ . , .. ~ . , .. ~ . r •••;~.; ::.;:L~ .r • ';::::.: ~.:': : ::: ."."•••::;;::-••:.~::.:..
-.,
'-"-.,
..
;:,~
;~~
;',j
;',j
:::::~.~::-..
-
:~.-.-,..-.,;:.
s~
::':-j
::':-j
j~:
i
..
,
4%
2%
0%
.....
s~
:------------------------
10%
:--------------------------
j~:
.....
,
.-------------------------
.-.:.-------------------~:.:
.......
,
.<
.
;:..:...;~-
............••••••./ .•,••\ ••,•.. •.•. .•. .•. .•.:::.
--r·-·,·-~-,-·-·,·-·T·-~,·-·-,-·-·,~:::/,:i~G·.-/:::·:':';:;:.::::.:::.: :.:.::..;':. "....,)'-:~~
:~~
(:
C·)
(:
C·)
")
")
Cohort Starting Date
c·
C:
,~
;
e.;;
C:
,~
;
(f ~
~
c::. ~
,~
c::.
;
,~
;
.-
':,
Cohort Starting Date
Int'I SF ex COO & other SF
other SF
8%,-----------------------------7% +-7,. ---------------------------
7%,------------------------------
6%
5%
5%
: '....
:
':\":.
':------------------,
: :
4%
•
':/:
.--------------------:
3%
......- - -
;:o2~~-;:~
..
;-\~
-:::
-:::
;\~
;\~
c.'
;~:
~:~ ::
;~:
:)
:..
"'1
,.\'1
:..
:..
;:;j
Cohort Starting Date
:)
;:;j
6%+-------~·~~-------------------
+-------~:-.:.---------------------
:~ :=============:=====~-:[U
.-------+--------:-~'-~::---'~
:~.: -:"~
2%
1%
0% ..
"\:'
. ",:.
:.•c...
~.I .. ~ ..I..I..,..I..,. . I..,. .I.. I..,{[:.:I ..'..I..'..I..'..I..,d ..'.. I.::'~
r.••,••
:)
;:;j
•. :•. :: • ..,. : •.
0.
Co
:':,.1
,',;
,~.
,',;
Cohort Starting Date
Note: Upgrades are measured on the alpha-numeric (or modified) rating scale. Rating cohorts are formed each month
covering a 12-month period. The latest 12-month cohort is formed on July 1, 2008 .
•~.~.~.:::.~~:l~~.~.f·December 2009 :':': Special Comment ::':: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Figure 9: Percentages of Downgrades (yel/ow line) and Upgrades (blue line) Preceded
by Watchlist Actions in the Same Direction
Global Structured Finance
Global Structured Finance ex SF COOs, other SF,
100%
80%
,A~~~LXHll~~~~~~~-----
t---~
60%
40%
,.",
.-".:
'-~:
;-\~
'-~:
.-".:
c:
,~
~"
,:;.
~,
c(
~5
~~
,<)
c:
,~
-,
,~
....:
-,
~
.0
::.\
Cohort Starting Date
USABSex HB.
~
60%
;~:
?£?
,:.:
.~j
USCMBS
it
C
,0;;'
-----------l
l
;-")
'-.:)
80%
::
60%
··.~1:.---r
'------------.
40%
-.
:-.~
g~
.-
~~
C
~~
~~;
'-....
,0;;'
._)
._)
40%
.......-.}
":':~:
2~~
:D
~~
~~;
.::
~
,y~
:;!I J-/------"~
60%
40%
20%
0%
~:'"
'C.
::/
';}
...~>
-~~
....... :
.-".:
.::
."
0
Cohort Starting Date
co:::
~"
c(
c:
:~
-~~
t
Cohort Starting Date
us HB. ex '05-'07 Vintages
100%
r9
---j;.-
80%
60%
40%
------------'·?:i:.
20%
00/0
·1··'··'··T··'··T-'-""·~·'··,··r·l··'··~·'··'··I·'~~~~~""",,\.
,,o..."j
~"
.,~
::::: Special Comment
··'··I~,·,-,-t·;·~~-,";"~~~-'·f··i"'"(~·r~··'··~·
C"'
••. . : ·----------;(:l,
80%
':\._._.t::';:'\. \
'c_
USHB.
100%
~:-
____:\;.:'.;-:. •:L.···:·i
20%
0%
Cohort Starting Date
~~~~~~~:::~!;~::~~~~f'December 2009
(;,
~~
100%
80%
0%
--
,~:
Cohort Starting Date
100%
20%
~~;
:)
:..
,0
.':)
:-'::
~~
j~:
;:;j
Cohort Starting Date
Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Figure 9 (continued): Percentages of Downgrades (yel/ow line) and Upgrades (blue
line) Preceded by Watchlist Actions in the Same Direction
us RM BS ex '05-'07 Vintages
us RMBS
100% - , - - - - -:
100%-,-----~
80%
+----
60%
+----:
"},}:~~ .
80%
+----
60%
+----:
40%
40%
20%
20%
CC,
.
;;:;
'A.'
S:,
8;
;c
t
;;:;
(~
;t.'
;c
;t.'
~
~
C
C
C·)
:.")
C'·
C
c
,0;;'
'-;:.
~~
~~;
80%
..f· .....;,
t0
n----"j'------.
.....
':: " "'"")u,""'
,-,",
...~>
....... :
~;j
"
".:
I··
C,~~
-,
....
~
-.
."
t
")
c',
,<)
oJ
t
,<)
\
,-,'\
.f
'-.:
:;\
i;;
,~:
r.'.J
1.-:
~~
80%
80%
60%
60%
40%
40%
20%
20%
....... :
~;j
"
".:
I··
C,~~
-,
....... -.
."
t
:::(
:-~:
::)
,:.:
.~j
'"
I..!-,;
C~
:::;
(:
~:
C·)
!s'
'"
other SF
100%
,-,',
t. r.' ..'..,;-· ·~l~~~-,---~~~
Cohort Starting Date
Int'I SFex COO & other SF
...~>
-.rf~~4----f\
+----'
0% -, .... ' .•.. f.' ....
100%
.Y,
,y~
. ::..):....>:,:
Cohort Starting Date
0%
.::
+-------:),---c:~__:---:~~.//::::.-.:
::: +-----:.•
20%
.Y,
'.'
Global COOs e x SF COOs
100% - , - - - - - - - - - - - - - - - -
.-:
..... L;. .:.--
+------::\
~
:~-(
Cohort Starting Date
Global COOs
:::
,'.
")
100% - , - - - - - - - - - - - - - - - - - ...
'0
~~~
:i;
Cohort Starting Date
80%
:~~t~l: ,"
';.',
".:)
;;:;
C·)
':.
0%
,<)
,-,'\
")
c',
oJ
t
Cohort Starting Date
,<)
'-.:
.f
:;\
i;;
,~:
1.-:
:::(
~~
:-~:
::)
,:.:
.~j
'"
I..!-,;
C~
~:
!s'
:::;
(:
C·)
'"
Cohort Starting Date
Note: At the beginning of each month, all securities that experienced downgrades (upgrades) within the next 12-month
period are grouped together to form a rating cohort. The latest 12-month cohort is formed on July 1, 2008. Gaps in the data
indicate that there were no downgrades (upgrades) during that time period.
~~~~~~~:::~~:i;::~~~~~~·.·December 2009
,:::: Special Comment ::::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Appendix: Description of Data Sample and Glossary
The data sample used in this report includes all public, 144A, and private tranches with a published Moody's
long-term global debt rating among global asset-backed securities (ABS), commercial and residential
mortgage-backed securities (CMBS and RMBS), collateralized debt obligations (COOs), and other structured
finance, including asset backed commercial paper (ABCP), structured investment vehicles (SIVs), structured
covered bonds, catastrophe bonds, and derivative product companies. Provisional ratings, credit estimates or
evaluations, short-term ratings, and national scale ratings are not included. The following types of securities
are excluded from the definition of global structured finance and therefore are not included in the data sample:
repackaged securities, structured notes, and other credit derivatives which are basically pass-throughs of the
rating of another entity.
This data set is an expansion of the data set that was used in prior structured finance performance studies.
In particular, this data sample:
7
::::
Includes tranches wrapped by financial guarantors, government agencies, and government sponsored
enterprises (GSEs);
::::
Includes interest-only (10) and residual tranches;
:.:.
Includes some transactions outside of the four major sectors (ABS, COO, CMBS, RMBS) of structured finance,
such as ABCP, SIVs, structured covered bonds, catastrophe bonds and derivative product companies;
:':'
Does not collapse tranches with the same rating from the same deal, i.e. all pari passu tranches are
counted in the data sample. The exceptions to this are notes with the same rating issued out of the same
program for ABCP, SIVs and structured covered bonds, in which case only the rating of the program and
not each individual security is counted.
The data used to create this report are commercially available via Moody's Structured Finance Default Risk service.
For more information, please email DefaultResearch@moodys.com.
.~.~.~~~.~.~y.........................................................................................................................................................................................................
Payment Shortfall
Structured finance securities are defined as having a payment shortfall (previously called "payment default") if
they have experienced either one of the following:
::::
Interest shortfall, or
::::
Principal write-down/loss.
Reductions in interest paid that arise due to prepayments of principal on the underlying loans or due to
limitations imposed by "available funds caps" (AFC) are not considered to be interest shortfalls. On the other
hand, "payment-in-kind" (PIK) events, in which the interest payment is deferred and capitalized into the
balance, are treated as interest shortfalls, regardless of whether or not it is described as a default event in the
bond's indenture. Explicit principal write-downs are included whereas implicit principal write-downs or undercollateralizations are not.
Material Impairment
Structured finance securities are defined as being in material impairment if they have:
7
:.:.
Sustained a payment shortfall that has not been cured, or
.:.:
Been downgraded to Ca or C, and hence is expected to suffer a significant level of payment losses in the
future.
The expanded data sample was first introduced in our 2007 rating transitions studies.
~~~~~!:i:~~::~::~~~~f·.December 2009
:':': Special Comment ':':' Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The impairment status of a security may change as it goes from cured (i.e. all outstanding shortfalls and losses
were repaid in full) to uncured (i.e. positive interest shortfalls or principal losses outstanding), or vice versa. If
a security downgraded to Ca or C, but not in payment shortfall, is subsequently upgraded, then it is no longer
in material impairment. Securities downgraded to Ca or C that are not upgraded are in material impairment
even if their payment shortfalls have been cured. Finally, securities with very minor shortfalls or losses are
excluded.
Principal Impairment
This refers to materially impaired securities that have experienced principal write-downs or principal losses, or
have been downgraded to Ca or C even if a principal write-down or loss has not yet been observed. In
particular, if a security has experienced principal write-down/loss or was downgraded to Ca or C, it is called a
principal impairment regardless of whether it has experienced interest shortfalls.
Interest Impairment
This refers to materially impaired securities that have experienced only interest shortfalls, no principal losses,
and were not downgraded to Ca or C.
Investment-Grade (IG) and Speculative-Grade (SG) Ratings
Investment-grade ratings refer to Aaa, Aa1, Aa2, Aa3, A1, A2, A3, Baa1, Baa2, and Baa3. Below investmentgrade or speculative-grade ratings refer to Ba1, Ba2, Ba3, B1, B2, B3, Caa1, Caa2, Caa3, Ca, and C.
Loss Severity or Loss-Given-Default (LGD)
The LGD rate of an impaired structured finance security is measured as the sum of the present values of net
losses, including both interest shortfalls and principal losses, discounted by the security's coupon rate and
expressed as a percentage of a given principal balance such as the principal balance at origination, at the
impairment date, or at any given cohort date.
Accuracy Ratio (AR)
An accuracy ratio based on the loss experience of structured finance securities (or an accuracy ratio adjusted
for loss-given-default, or LGD) is the ratio of the area between a loss-based cumulative accuracy profile (CAP)
curve and the 45-degree line to the maximum possible area above the 45-degree line. The loss-based CAP
curve (or a CAP curve adjusted for LGD) plots, for each rating category, the proportion of the losses of all
impaired securities accounted for by securities with the same or lower rating against the proportion of all
securities in the sample population with the same or lower rating. 8
To calculate accuracy ratios, rating cohorts are formed for each calendar month over the study period so that
all outstanding securities in the sample population are grouped together by their alpha-numeric ratings at the
beginning of the month. For each monthly rating cohort, we determine the number of securities that became
impaired within one or five years of the cohort formation date and their loss severity rates as a percentage of
the cohort date balance. Cumulative shares of securities rank-ordered by rating are calculated for the
universe of all securities and the universe of impaired securities, respectively, and based on this, a CAP curve
is plotted. Note that only LGD for principal impaired securities are used in the calculation. Please see the
definition of LGD for further details.
The CAP curve adjusted for LGD is also known as a "power curve" because it shows how effective a rating
system is at differentiating between securities that have sustained high losses from securities that have
sustained low or no losses. The metric is defined relative to the distribution of ratings in the population. The
accuracy ratio is an effective way to summarize the CAP curve into a single number. The accuracy ratio is
zero if the CAP curve collapses to the 45-degree line, suggesting that all impaired securities are randomly
distributed throughout the population without regard to rating.
8
For an illustration of the CAP curve adjusted for LGD, see "Default & Loss Rates of Structured Finance Securities: 1993-2003," Moody's
Special Comment, September 2004.
~~~~~!:i:~~~:::~~~~~f·.December 2009
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Investment-Grade Loss Rate
The one-year investment-grade loss rate is calculated as follows. First, for a given cohort, we compute the
LGD as a share of the tranche balance as of the cohort date for each security that carried an investment-grade
rating at the cohort formation date and became impaired within a 12-month period after the cohort date. We
then take the sum of these LGD rates and divide by the total number of investment-grade securities
outstanding as of the cohort formation date. Note that the LGD rate is not weighted by dollar volume and that
only LGD for principal impaired securities are used in the calculation. The five-year investment-grade loss rate
is calculated similarly.
Average Rating Before Impairment
The rating of an impaired security is measured every month for 36 months prior to impairment. These 36
rating measurements are averaged together to create one representative number for each impaired security.
For a particular cohort, the average rating before impairment is the weighted average of these average ratings
for each security that became impaired within 12 months after the cohort formation date. The weight for each
security is the LGD rate of the tranche as a share of its original balance. This weighting scheme will place
greater emphasis on the average ratings of impaired tranches with higher LGD over impaired tranches with
lower LGD. 9 Note that only LGD for principal impaired securities are used in the calculation.
Rating Action Rate (Downgrade and Upgrade Rate)
The rating action rate is defined as the number of securities that experienced a rating change within a year
after cohort formation divided by the total number of securities outstanding at the cohort formation date (the
beginning of each month). Rating changes are measured on the alpha-numeric rating scale and are based on
comparing the rating at the beginning and end of the time period under consideration. However, if a rating
was withdrawn by the end of the time period, then the rating prior to withdrawal is used as the end rating.
Note that a security will only be counted if it was outstanding as of the cohort formation date. Downgrade
rates and upgrade rates are measured similarly based on downgrade and upgrade rating actions, respectively.
Large Rating Action Rate
A large ration action is said to occur if a rating action (or cumulative rating actions) cause(s) a security's rating
to change by three or more notches within a year after cohort formation. The large rating action rate is the
number of such securities divided by the total number of securities outstanding at the cohort formation date.
Large downgrade rates and large upgrade rates are measured similarly based on large downgrade and large
upgrade rating actions, respectively.
Percentage of Downgrades (Upgrades) Preceded by Watchlist Actions
in the Same Direction
This metric is defined as the total number of downgraded (upgraded) securities that were placed on the
watchlist in the same direction before they were downgraded (upgraded), divided by the total number of
securities that were downgraded (upgraded) within 12 months after the cohort formation date.
ABS ex HEL
ABS stands for asset-backed securities. This structured finance sector includes securities backed by both
traditional asset types such as auto loans, credit card receivables, student loans, and manufactured housing
loans, and non-traditional asset types such as mutual fund fees, tax liens, tobacco settlement payments, and
intellectual property. Home equity loans (HEL) are explicitly excluded from US ABS ex HEL.
9 We began using LGD rates as weights in computing the average rating before impairment in the full-year 2005 structured finance
performance report. Ideally, LGD rates should be calculated as a percentage of the principal balance outstanding for each month in the 36
months prior to the impairment, and ratings should then be weighted by these monthly LGD rates. Practically, however, it does not make a
substantial difference in the average rating number whether we use LGD rates as a share of impairment-date balance, original balance, or
monthly principal outstanding. Since the monthly LGD rates are very time consuming to compute due to amortization, we use the LGD rate
as a share of original balance as the weight variable.
~~~~~!:i:~!:i:~~~~~f·.December 2009
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
HEL
The home equity loan or HEL sector includes securities backed by subprime (B&C) mortgage loans, home
improvement loans, high loan-to-value (high LTV) loans, home equity lines of credit (HELOCs), and closedend second-lien loans, as well as net interest margin (NIM) securitizations. It does not include securities
backed by Alt-A mortgages, which are included in the RMBS sector. HEL is part of the ABS sector.
Prior to 1998, RMBS collateral was generally defined as first-lien residential mortgages, regardless of the
credit quality of the borrower. HEL collateral generally included junior liens such as HELOCs or closed-end
seconds. However, as subprime lending became more prevalent, the market shifted its definition such that
HEL encompassed subprime first-lien residential mortgages while RMBS included first-lien mortgages made to
higher quality borrowers. Since 1998, a deal classified as RMBS by Moody's is generally backed by prime or
Alt-A quality first-lien residential mortgages, while a deal classified as HEL is generally backed by subprime
first-lien mortgages or junior liens. Therefore, a subprime deal which would be classified as HEL today may
have been classified as RMBS in the past.
RMBS
RMBS stands for residential mortgage-backed securities. The vast majority of these securities are backed by
first-lien prime mortgages or by Alt-A mortgages. For further details, see the definition of HEL.
CMBS
CMBS stands for commercial mortgage-backed securities. Commercial real estate (CRE) COOs, where 70%
or more of the collateral is comprised of CRE loans, are classified as CMBS. If the collateral backing the
transaction contains less than 70% CRE loans, then the deal is classified as a COO.
COOs
COOs stand for collateralized debt obligations. Derivative securities such as structured notes and repackaged
securities are not considered to be part of this sector. Commercial real estate (CRE) COOs are also excluded
(see the definition of CMBS).
Other Structured Finance
Other structured finance consists of structured finance securities not categorized in the five major sectors
(ABS ex HEL, HEL, RMBS, CMBS, and COO) including asset-backed commercial paper (ABCP) programs,
structured investment vehicles (SIVs), structured covered bonds, insurance-linked securities such as
catastrophe bonds, and derivative product companies. However, notes carrying only short-term ratings such
as commercial paper are excluded.
Global Structured Finance
Global structured finance captures securities issued around the world in the five major sectors - ABS ex HEL,
HEL, RMBS, CMBS, and COO - and in the Other Structured Finance category.
US Structured Finance
US structured finance securities are denominated in US dollars and issued in the US market or denominated
in Canadian dollars and issued in Canada. In cases where the source of the underlying collateral and the
denomination of the securities cross multiple countries/regions, deals are classified by the location at which
they are monitored.
Inti SF ex COO and Other SF
This refers to securities that are not denominated in US dollars and issued in the US market and not
denominated in Canadian dollars and issued in Canada. The majority of the securities in this sector are
issued in Europe, the Middle East, and Africa (EMEA); the rest are issued in the Asia Pacific region and Latin
America. COOs and Other SF are excluded.
~~~~~!:i:~~::~::~~~~f·.December 2009
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Moody's Related Research
Special Comments:
,:,:
The Performance of Structured Finance Ratings: Full-Year 2008 Report, August 2009 (119780)
:,:,
The Performance of Structured Finance Ratings: Mid-Year 2008 Report, November 2008 (112347)
,:,:
The Performance of Structured Finance Ratings: Full-Year 2007 Report, July 2008 (110179)
:,:,
The Performance of Structured Finance Ratings: Mid-Year 2007 Report, October 2007 (105390)
':':
The Performance of Structured Finance Ratings: Full-Year 2006 Report, May 2007 (103017)
:':'
The Performance of Structured Finance Ratings: Mid-Year 2006 Report, September 2006 (99034)
':':
The Performance of Structured Finance Ratings: Full-Year 2005 Report, May 2006 (97346)
:':'
The Performance of Structured Finance Ratings: Mid-Year 2005 Report, September 2005 (94463)
':':
Default & Loss Rates of Structured Finance Securities: 1993-2008, August 2009 (119617)
:':'
Measuring Loss-Given-Default for Structured Finance Securities: An Update, December 2006 (101284)
':':
Structured Finance Rating Transitions: 1983-2008, March 2009 (115157)
:':'
Japanese Structured Finance Rating Transitions: 1994-2008, March 2009 (115070)
':':
Asia-Pacific (ex-Japan) Structured Finance Rating Transitions: 1990-2008, March 2009 (115165)
:':'
EMEA Structured Finance Rating Transitions: 1988-2008, April 2009 (116507)
':':
Deal Sponsor and Credit Risk of U,S, ABS and MBS Securities, December 2006 (100872)
:':'
The Relationship between Par Coupon Spreads and Credit Ratings in US Structured Finance, December
2005 (95494)
,:,:
Structured Finance Watchlist Resolutions: 1992-2003, June 2004 (87305)
:,:,
The Performance of Moody's Corporate Debt Ratings: September 2009 Quarterly Update, October 2009
(120706)
:':'
Measuring the Performance of Corporate Bond Ratings, April 2003 (77916)
':':
Guide to Moody's Default Research: October 2009 Update, October 2009 (120969)
To access any of these reports, click on the entry above, Note that these references are current as of the date of publication
of this report and that more recent reports may be available, All research may not be available to all clients,
~~~~~!:i:~~:::::~~~~f',December 2009
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
The Performance of Structured Finance Ratings: Mid-Year 2009 Report
Report Number: 121626
Julia Tung
Geraldine Kim
Sylviane Grant
© Copyright 2009, Moody's Investors Service, Inc. and/or its licensors and affiliates including Moody's Assurance Company, Inc. (together, "MOODY'S"). All rights
reserved. ALL INFORMATION CONTAINED HEREIN IS PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR
OTHERWISE REPRODUCED, REPACKAGED, FURTHER TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED
FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN WHOLE OR IN PART, IN ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER, BY ANY
PERSON WITHOUT MOODY'S PRIOR WRITTEN CONSENT. All information contained herein is obtained by MOODY'S from sources believed by it to be accurate
and reliable. Because of the possibility of human or mechanical error as well as other factors, however, such information is provided "as is" without warranty of any
kind and MOODY'S, in particular, makes no representation or warranty, express or implied, as to the accuracy, timeliness, completeness, merchantability or fitness
for any particular purpose of any such information. Under no circumstances shall MOODY'S have any liability to any person or entity for (a) any loss or damage in
whole or in part caused by, resulting from, or relating to, any error (negligent or otherwise) or other circumstance or contingency within or outside the control of
MOODY'S or any of its directors, officers, employees or agents in connection with the procurement, collection, compilation, analysis, interpretation, communication,
publication or delivery of any such information, or (b) any direct, indirect, special, consequential, compensatory or incidental damages whatsoever (including without
limitation, lost profits), even if MOODY'S is advised in advance of the possibility of such damages, resulting from the use of or inability to use, any such information.
The credit ratings and financial reporting analysis observations, if any, constituting part of the information contained herein are, and must be construed solely as,
statements of opinion and not statements of fact or recommendations to purchase, sell or hold any securities. NO WARRANTY, EXPRESS OR IMPLIED, AS TO
THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR
OTHER OPINION OR INFORMATION IS GIVEN OR MADE BY MOODY'S IN ANY FORM OR MANNER WHATSOEVER. Each rating or other opinion must be
weighed solely as one factor in any investment decision made by or on behalf of any user of the information contained herein, and each such user must accordingly
make its own study and evaluation of each security and of each issuer and guarantor of, and each provider of credit support for, each security that it may consider
purchasing, holding or selling.
MOODY'S hereby discloses that most issuers of debt securities (including corporate and municipal bonds, debentures, notes and commercial paper) and
preferred stock rated by MOODY'S have, prior to assignment of any rating, agreed to pay to MOODY'S for appraisal and rating services rendered by it fees ranging
from $1,500 to approximately $2,400,000. Moody's Corporation (MCa) and its wholly-owned credit rating agency subsidiary, Moody's Investors Service (MIS), also
maintain policies and procedures to address the independence of MIS's ratings and rating processes. Information regarding certain affiliations that may exist
between directors of MCa and rated entities, and between entities who hold ratings from MIS and have also publicly reported to the SEC an ownership interest in
MCa of more than 5%, is posted annually on Moody's website at www.moodys.com under the heading "Shareholder Relations - Corporate Governance - Director
and Shareholder Affiliation Policy."
Moody's Investors Service
~~~~~!:i:~~~:~::~~~~f'December 2009
::::: Special Comment ':::: Moody's Credit Policy - The Performance of Structured Finance Ratings: Mid-Year 2009 Report
@FedFRASER